Files
akbasic/docs/04-control-flow.md

421 lines
9.0 KiB
Markdown
Raw Normal View History

# 4. Control flow
## IF ... THEN ... ELSE
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 A# = 5
20 IF A# = 5 THEN PRINT "FIVE" ELSE PRINT "NOT FIVE"
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
FIVE
```
The condition is a whole expression, so `AND`, `OR` and `NOT` all work in one:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 A# = 5
20 IF A# > 0 AND A# < 10 THEN PRINT "IN RANGE"
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
IN RANGE
```
**Everything after `THEN` on the line belongs to the condition**, which matters as soon
as you put several statements on a line:
| Line | Condition | What runs |
|---|---|---|
| `IF C THEN A : B` | true | `A`, `B` |
| `IF C THEN A : B` | false | nothing |
| `IF C THEN A ELSE B : D` | true | `A` |
| `IF C THEN A ELSE B : D` | false | `B`, `D` |
The remainder always belongs to whichever arm was written *last*.
### Blocks
`BEGIN` and `BEND` make an `IF` span lines:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 A# = 5
20 IF A# = 5 THEN BEGIN
30 PRINT "IN THE BLOCK"
40 PRINT "STILL IN IT"
50 BEND
60 PRINT "AFTER"
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
IN THE BLOCK
STILL IN IT
AFTER
```
When the condition is false every line up to the `BEND` is skipped.
## FOR ... NEXT
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 FOR I# = 1 TO 5
20 PRINT I#
30 NEXT I#
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
1
2
3
4
5
```
`STEP` sets the stride, and a negative one counts down:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 FOR I# = 10 TO 0 STEP -2
20 PRINT I#
30 NEXT I#
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
10
8
6
4
2
0
```
The counter is an ordinary variable and the body may assign to it. Two things to know:
- **The counter does not survive the loop.** It lives in the loop's own scope, so
reading it afterwards gives zero. On a C128 it keeps its final value.
- **A step that overshoots runs the body one extra time.** `FOR I = 1 TO 9 STEP 3` runs
with 1, 4, 7 *and 10*. Both are recorded as known defects; see Chapter 13.
`EXIT` leaves the loop early:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 FOR I# = 1 TO 100
20 IF I# = 5 THEN EXIT
30 NEXT I#
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
40 PRINT "OUT"
```
```output
OUT
```
## DO ... LOOP
The condition can go on either end, or neither:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 I# = 0
20 DO WHILE I# < 3
30 PRINT I#
40 I# = I# + 1
50 LOOP
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
0
1
2
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 I# = 0
20 DO
30 PRINT I#
40 I# = I# + 1
50 LOOP UNTIL I# = 3
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
0
1
2
```
A condition on the `DO` is tested before the body, so the body may run zero times. A
condition on the `LOOP` is tested after, so it runs at least once. `DO` with no
condition at all loops forever until an `EXIT` or a `GOTO` leaves it.
`EXIT` works here too.
Implement generators: GEN, EMIT, END GEN, FOR EACH and DO EACH (issue 57) Adds generator support per the plan in issue 57: - environment.h: isGenerator/generatorFn on a GEN call's own environment, isEachLoop and forGeneratorEnv on a FOR EACH/DO EACH loop's own environment. - runtime.c: splits akbasic_runtime_prev_environment() into akbasic_runtime_detach_environment() (return to parent without releasing) and akbasic_runtime_release_environment() (give variables and the pool slot back, on any environment); prev_environment() is now the two in sequence. akbasic_runtime_call_function() refuses to call a GEN like an ordinary function. - verbs.c/verbs.h/scanner: new keywords GEN, EMIT, EACH, IN and the compound verb "END GEN" (built by a new akbasic_parse_end(), the same trick akbasic_parse_print() uses for PRINT #). - parser_commands.c: akbasic_parse_gen() (modelled on multi-line DEF), akbasic_parse_end(), and EACH branches in akbasic_parse_for()/ akbasic_parse_do(). - runtime_generator.c (new): akbasic_cmd_gen, akbasic_cmd_emit, akbasic_cmd_end_gen, and the invoke/pump/release machinery FOR EACH, DO EACH, NEXT and LOOP share. EMIT walks up to the nearest isGenerator ancestor rather than assuming it is standing directly in the GEN's own call frame, because a GEN body may nest its own FOR/DO/GOSUB around an EMIT -- the issue's own ROOMOBJECTS example does exactly that. - runtime_commands.c/runtime_structure.c: EACH branches in cmd_for/cmd_do, matching EACH branches in cmd_next/cmd_loop, and forGeneratorEnv release on every path that can abandon a live generator (EXIT, a NEXT that pops for a mismatched loop variable). Deviates from the plan in one place: FunctionDef gained an isGenerator flag (not in the plan's field list) because refusing a GEN called like a function has to happen before anything is pushed. Relying on EMIT's own isGenerator check for that case doesn't work: akbasic_runtime_call_function() drives its own step loop the same way akbasic_runtime_pump_generator() does, and a BASIC-level error inside that loop is swallowed by process_line_run() as reported-but-not-propagated, so the call would silently "succeed" with a meaningless return value instead of failing. Also: a zero-argument parameter list is not supported by the DEF/GEN parameter parser this reuses (a pre-existing limitation, not generator-specific); every generator in the tests takes at least one parameter as a result. Tests: tests/generators.c (pool exhaustion under repeated EXIT, calling a GEN like a function, EMIT outside a GEN, self-recursion, sibling/nested invocations) and tests/language/flowcontrol/generators_*.bas -- the issue's own ROOMOBJECTS example in both loop shapes, an empty generator, non-numeric EMIT, nested/interleaved invocations, and three error-path golden cases. Docs: control-flow chapter 4 gets a GEN/EMIT/FOR EACH/DO EACH section, the verb reference gets GEN/EMIT/END GEN entries and updated FOR/DO/NEXT/LOOP/EXIT rows, and architecture chapter 14 documents the detach/release split and the two-environment generator invocation. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-08-06 10:02:43 -04:00
## GEN ... EMIT and FOR EACH / DO EACH
A `GEN` is a subroutine that hands back more than one value, one at a time, instead of
returning once. It looks like a multi-line `DEF`, except its body runs `EMIT` where a
function would `RETURN`, and it ends in `END GEN` rather than `RETURN`:
```basic
10 GEN COUNTUP(N#)
20 FOR I# = 1 TO N#
30 EMIT I#
40 NEXT I#
50 END GEN
60 FOR EACH V# IN COUNTUP(3)
70 PRINT V#
80 NEXT V#
```
```output
1
2
3
```
`FOR EACH` is what runs a `GEN`: it calls `COUNTUP(3)` the way `FOR EACH V# IN` says,
and each `EMIT` inside the generator's body becomes one trip through the loop, with `V#`
holding whatever was emitted. When the generator's body reaches `END GEN` with nothing
left to emit, the loop ends -- there is no separate "no more values" check to write.
`DO EACH ... LOOP` does the same thing:
```basic
10 GEN COUNTUP(N#)
20 FOR I# = 1 TO N#
30 EMIT I#
40 NEXT I#
50 END GEN
60 DO EACH V# IN COUNTUP(3)
70 PRINT V#
80 LOOP
```
```output
1
2
3
```
A `GEN`'s body is ordinary BASIC: it may hold its own `FOR`, `DO`, `IF` or `GOSUB`
around the `EMIT`s, and even invoke another `GEN` with its own `FOR EACH`/`DO EACH` --
the same generator invoked with different arguments, nested or side by side, is not
recursion. A `GEN` invoking *itself* from within its own currently-running body is
Fix generator teardown leaks, add RETURN-in-GEN and LOOP conditions on DO EACH Review findings and follow-ups from PR #61 review: - runtime_generator.c: akbasic_runtime_release_generator() now releases the forGeneratorEnv of every scope it walks through. Abandoning a generator that was itself suspended inside a FOR EACH over another generator stranded the inner generator's pool slot; a loop doing so exhausted the twelve-slot pool and died far from the cause. - runtime.c/runtime.h: new akbasic_runtime_unwind_to_environment(), the shared teardown for the error unwinds in pump_generator() and call_function() -- both previously bare prev_environment() loops with the same suspended-generator blindness. - runtime_commands.c: bare RETURN standing in a GEN's own frame ends the generator exactly as END GEN does -- a GEN is a function at heart. RETURN with a value there is refused (values leave a GEN only through EMIT). The no-frame error message now says "GOSUB, DEF, or GEN". - runtime_structure.c: LOOP WHILE/UNTIL composes with DO EACH -- checked after each trip with the loop variable still holding that trip's value; a condition that stops the loop abandons the generator exactly as EXIT does. Previously the condition was silently ignored, while the verb reference documented it as working. - parser_commands.c: trailing tokens after the generator call on a FOR EACH/DO EACH line are refused at parse. Previously they sat unparsed and blew up only after the loop completed, when the parent scope resumed the line mid-statement -- an error at the loop's end pointing at its start. - tests/generators.c: pool-exhaustion tests for the nested-abandonment and LOOP-condition paths, RETURN semantics tests, and a direct test of the unwind primitive. Three new golden pairs cover RETURN, LOOP conditions and the misplaced-condition parse error. - docs: RETURN and LOOP-condition semantics in 04-control-flow.md and 11-verb-reference.md; corrected the self-recursion analogy (functions are re-entrant here). TODO.md 1.10 records the generator design decisions the code comments were already citing, plus the zero-arg parameter-list limitation. MAINTENANCE.md gains the abandoned-generators invariant those comments also cited. Co-Authored-By: Andrew Kesterson <andrew@aklabs.net> Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-06 11:29:17 -04:00
refused: unlike a function call, which runs to completion and returns, the outer
invocation is suspended mid-body waiting on the same loop, and there is no answer to
"which EMIT feeds which loop" that is not a surprise.
Implement generators: GEN, EMIT, END GEN, FOR EACH and DO EACH (issue 57) Adds generator support per the plan in issue 57: - environment.h: isGenerator/generatorFn on a GEN call's own environment, isEachLoop and forGeneratorEnv on a FOR EACH/DO EACH loop's own environment. - runtime.c: splits akbasic_runtime_prev_environment() into akbasic_runtime_detach_environment() (return to parent without releasing) and akbasic_runtime_release_environment() (give variables and the pool slot back, on any environment); prev_environment() is now the two in sequence. akbasic_runtime_call_function() refuses to call a GEN like an ordinary function. - verbs.c/verbs.h/scanner: new keywords GEN, EMIT, EACH, IN and the compound verb "END GEN" (built by a new akbasic_parse_end(), the same trick akbasic_parse_print() uses for PRINT #). - parser_commands.c: akbasic_parse_gen() (modelled on multi-line DEF), akbasic_parse_end(), and EACH branches in akbasic_parse_for()/ akbasic_parse_do(). - runtime_generator.c (new): akbasic_cmd_gen, akbasic_cmd_emit, akbasic_cmd_end_gen, and the invoke/pump/release machinery FOR EACH, DO EACH, NEXT and LOOP share. EMIT walks up to the nearest isGenerator ancestor rather than assuming it is standing directly in the GEN's own call frame, because a GEN body may nest its own FOR/DO/GOSUB around an EMIT -- the issue's own ROOMOBJECTS example does exactly that. - runtime_commands.c/runtime_structure.c: EACH branches in cmd_for/cmd_do, matching EACH branches in cmd_next/cmd_loop, and forGeneratorEnv release on every path that can abandon a live generator (EXIT, a NEXT that pops for a mismatched loop variable). Deviates from the plan in one place: FunctionDef gained an isGenerator flag (not in the plan's field list) because refusing a GEN called like a function has to happen before anything is pushed. Relying on EMIT's own isGenerator check for that case doesn't work: akbasic_runtime_call_function() drives its own step loop the same way akbasic_runtime_pump_generator() does, and a BASIC-level error inside that loop is swallowed by process_line_run() as reported-but-not-propagated, so the call would silently "succeed" with a meaningless return value instead of failing. Also: a zero-argument parameter list is not supported by the DEF/GEN parameter parser this reuses (a pre-existing limitation, not generator-specific); every generator in the tests takes at least one parameter as a result. Tests: tests/generators.c (pool exhaustion under repeated EXIT, calling a GEN like a function, EMIT outside a GEN, self-recursion, sibling/nested invocations) and tests/language/flowcontrol/generators_*.bas -- the issue's own ROOMOBJECTS example in both loop shapes, an empty generator, non-numeric EMIT, nested/interleaved invocations, and three error-path golden cases. Docs: control-flow chapter 4 gets a GEN/EMIT/FOR EACH/DO EACH section, the verb reference gets GEN/EMIT/END GEN entries and updated FOR/DO/NEXT/LOOP/EXIT rows, and architecture chapter 14 documents the detach/release split and the two-environment generator invocation. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-08-06 10:02:43 -04:00
`EXIT` leaves a `FOR EACH`/`DO EACH` loop early, exactly as it does a plain `FOR` or
`DO`, and the generator it was consuming stops there -- nothing forces the rest of it to
run just because the loop started it:
```basic
10 GEN COUNTUP(N#)
20 FOR I# = 1 TO N#
30 EMIT I#
40 NEXT I#
50 END GEN
60 FOR EACH V# IN COUNTUP(100)
70 PRINT V#
80 IF V# = 3 THEN EXIT
90 NEXT V#
100 PRINT "STOPPED"
```
```output
1
2
3
STOPPED
```
Fix generator teardown leaks, add RETURN-in-GEN and LOOP conditions on DO EACH Review findings and follow-ups from PR #61 review: - runtime_generator.c: akbasic_runtime_release_generator() now releases the forGeneratorEnv of every scope it walks through. Abandoning a generator that was itself suspended inside a FOR EACH over another generator stranded the inner generator's pool slot; a loop doing so exhausted the twelve-slot pool and died far from the cause. - runtime.c/runtime.h: new akbasic_runtime_unwind_to_environment(), the shared teardown for the error unwinds in pump_generator() and call_function() -- both previously bare prev_environment() loops with the same suspended-generator blindness. - runtime_commands.c: bare RETURN standing in a GEN's own frame ends the generator exactly as END GEN does -- a GEN is a function at heart. RETURN with a value there is refused (values leave a GEN only through EMIT). The no-frame error message now says "GOSUB, DEF, or GEN". - runtime_structure.c: LOOP WHILE/UNTIL composes with DO EACH -- checked after each trip with the loop variable still holding that trip's value; a condition that stops the loop abandons the generator exactly as EXIT does. Previously the condition was silently ignored, while the verb reference documented it as working. - parser_commands.c: trailing tokens after the generator call on a FOR EACH/DO EACH line are refused at parse. Previously they sat unparsed and blew up only after the loop completed, when the parent scope resumed the line mid-statement -- an error at the loop's end pointing at its start. - tests/generators.c: pool-exhaustion tests for the nested-abandonment and LOOP-condition paths, RETURN semantics tests, and a direct test of the unwind primitive. Three new golden pairs cover RETURN, LOOP conditions and the misplaced-condition parse error. - docs: RETURN and LOOP-condition semantics in 04-control-flow.md and 11-verb-reference.md; corrected the self-recursion analogy (functions are re-entrant here). TODO.md 1.10 records the generator design decisions the code comments were already citing, plus the zero-arg parameter-list limitation. MAINTENANCE.md gains the abandoned-generators invariant those comments also cited. Co-Authored-By: Andrew Kesterson <andrew@aklabs.net> Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-06 11:29:17 -04:00
`RETURN` ends a `GEN` from the inside, exactly as it ends a multi-line `DEF` or a
`GOSUB`: the generator is done, the loop consuming it ends, and the program carries on
after the loop. What a generator's `RETURN` cannot do is carry a value -- values leave
a `GEN` one at a time, through `EMIT`, and `RETURN 99` inside one is an error. Like a
`GOSUB`'s or a `DEF`'s, the `RETURN` must stand in the `GEN`'s own scope: from inside
a `FOR` or `DO` the body opened, it is an error, though `IF ... THEN RETURN` is fine
because `IF` opens no scope of its own.
```basic
10 GEN FIRSTFEW(N#)
20 EMIT 1
30 IF N# < 2 THEN RETURN
40 EMIT 2
50 IF N# < 3 THEN RETURN
60 EMIT 3
70 END GEN
80 FOR EACH V# IN FIRSTFEW(2)
90 PRINT V#
100 NEXT V#
110 PRINT "DONE"
```
```output
1
2
DONE
```
A `DO EACH`'s `LOOP` may carry a `WHILE` or `UNTIL`, and the two compose: the condition
is checked after each trip through the body, with the loop variable still holding that
trip's value, and a condition that says stop abandons the rest of the generator exactly
as `EXIT` does. The condition belongs on the `LOOP` -- putting it on the `DO` line is
an error, since the `DO` line already says what the loop consumes.
```basic
10 GEN COUNTUP(N#)
20 FOR I# = 1 TO N#
30 EMIT I#
40 NEXT I#
50 END GEN
60 DO EACH V# IN COUNTUP(10)
70 PRINT V#
80 LOOP UNTIL V# = 3
90 PRINT "STOPPED"
```
```output
1
2
3
STOPPED
```
Implement generators: GEN, EMIT, END GEN, FOR EACH and DO EACH (issue 57) Adds generator support per the plan in issue 57: - environment.h: isGenerator/generatorFn on a GEN call's own environment, isEachLoop and forGeneratorEnv on a FOR EACH/DO EACH loop's own environment. - runtime.c: splits akbasic_runtime_prev_environment() into akbasic_runtime_detach_environment() (return to parent without releasing) and akbasic_runtime_release_environment() (give variables and the pool slot back, on any environment); prev_environment() is now the two in sequence. akbasic_runtime_call_function() refuses to call a GEN like an ordinary function. - verbs.c/verbs.h/scanner: new keywords GEN, EMIT, EACH, IN and the compound verb "END GEN" (built by a new akbasic_parse_end(), the same trick akbasic_parse_print() uses for PRINT #). - parser_commands.c: akbasic_parse_gen() (modelled on multi-line DEF), akbasic_parse_end(), and EACH branches in akbasic_parse_for()/ akbasic_parse_do(). - runtime_generator.c (new): akbasic_cmd_gen, akbasic_cmd_emit, akbasic_cmd_end_gen, and the invoke/pump/release machinery FOR EACH, DO EACH, NEXT and LOOP share. EMIT walks up to the nearest isGenerator ancestor rather than assuming it is standing directly in the GEN's own call frame, because a GEN body may nest its own FOR/DO/GOSUB around an EMIT -- the issue's own ROOMOBJECTS example does exactly that. - runtime_commands.c/runtime_structure.c: EACH branches in cmd_for/cmd_do, matching EACH branches in cmd_next/cmd_loop, and forGeneratorEnv release on every path that can abandon a live generator (EXIT, a NEXT that pops for a mismatched loop variable). Deviates from the plan in one place: FunctionDef gained an isGenerator flag (not in the plan's field list) because refusing a GEN called like a function has to happen before anything is pushed. Relying on EMIT's own isGenerator check for that case doesn't work: akbasic_runtime_call_function() drives its own step loop the same way akbasic_runtime_pump_generator() does, and a BASIC-level error inside that loop is swallowed by process_line_run() as reported-but-not-propagated, so the call would silently "succeed" with a meaningless return value instead of failing. Also: a zero-argument parameter list is not supported by the DEF/GEN parameter parser this reuses (a pre-existing limitation, not generator-specific); every generator in the tests takes at least one parameter as a result. Tests: tests/generators.c (pool exhaustion under repeated EXIT, calling a GEN like a function, EMIT outside a GEN, self-recursion, sibling/nested invocations) and tests/language/flowcontrol/generators_*.bas -- the issue's own ROOMOBJECTS example in both loop shapes, an empty generator, non-numeric EMIT, nested/interleaved invocations, and three error-path golden cases. Docs: control-flow chapter 4 gets a GEN/EMIT/FOR EACH/DO EACH section, the verb reference gets GEN/EMIT/END GEN entries and updated FOR/DO/NEXT/LOOP/EXIT rows, and architecture chapter 14 documents the detach/release split and the two-environment generator invocation. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-08-06 10:02:43 -04:00
A `GEN` shares its namespace with `DEF` -- the same name cannot be both -- but it is not
a function and cannot be called like one: `X# = COUNTUP(3)` is refused, because nothing
about an ordinary call means "resume where the last `EMIT` left off." `FOR EACH`/`DO
EACH` is the only thing that consumes a `GEN`.
## GOTO and GOSUB
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 GOSUB 100
20 PRINT "BACK"
30 END
100 PRINT "IN THE SUBROUTINE"
110 RETURN
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
IN THE SUBROUTINE
BACK
```
`RETURN` goes back to the line after the `GOSUB`.
## Labels
A label is a name with no type suffix. It marks a line, and anything that takes a line
number takes a label instead:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 GOTO SETUP
20 PRINT "SKIPPED"
100 LABEL SETUP
110 PRINT "ARRIVED"
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
ARRIVED
```
**Labels are filed before the program runs**, so a forward `GOTO` works. This is worth
using for its own sake: a program written with labels is immune to `RENUMBER`, because
there is no number to rewrite.
Follow that one step further and the numbers stop earning their keep at all. A program
in a file can leave them out — see
[Chapter 2](02-getting-started.md#a-program-in-a-file-does-not-need-them) — and then
every branch it makes is by name:
```basic
GOSUB GREET
GOTO FINISH
PRINT "NOT REACHED"
LABEL GREET
PRINT "HELLO"
RETURN
LABEL FINISH
PRINT "DONE"
```
```output
HELLO
DONE
```
Nothing in that program can be broken by inserting a line into it, which is the whole
argument.
## ON
`ON` picks the *n*th target from a list, counting from one:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 CHOICE# = 2
20 ON CHOICE# GOTO 100, 200, 300
30 PRINT "CHOICE WAS OUT OF RANGE"
40 END
100 PRINT "FIRST"
110 END
200 PRINT "SECOND"
210 END
300 PRINT "THIRD"
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
SECOND
```
Out of range is not an error — it falls through to the next statement, which is what
lets you write the check as the line after. `ON ... GOSUB` works the same way and
returns.
## Trapping errors
`TRAP` sends an error to a handler instead of stopping the program:
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```basic
10 TRAP HANDLER
20 DIM Q#(2)
30 PRINT Q#(9)
40 PRINT "CARRIED ON"
50 END
100 LABEL HANDLER
110 PRINT "CAUGHT " + ERR(ER#) + " ON LINE " + EL#
120 RESUME NEXT
```
Execute every documented example as a test docs/ and README.md carry 85 fenced blocks. Every one was checked by hand exactly once, when it was written, which is not a standard that survives a changing interpreter -- and four were already wrong: two transcripts showing a leading space PRINT does not emit, akbasic_TextSink in README.md missing the two members it had grown hours earlier, and FILTER's refusal quoted with wording the code does not use. tests/docs_examples.sh reads a fence-tag vocabulary and runs what it finds. BASIC programs and transcripts run and are byte-compared against an `output` block; C snippets compile with -fsyntax-only against the real include path, which CMake writes out because it is transitive through akerror, akstdlib and akgl; shell blocks run in a sandbox. Anything that would reconfigure the build tree, hit the network or re-enter the suite is tagged norun with the reason in MAINTENANCE.md, and the two cmake blocks stay hand-maintained by decision. An untagged block is a failure rather than a default, and the pass line reports what it executed by kind. Both exist because the way a harness like this dies is by quietly matching nothing and passing -- which it duly did on the first CTest run, where a generator expression evaluating to nothing still contributed an empty argument that the script read as a filename. The count is what caught it. The excerpt check earns its own mention: a block tagged `c excerpt=include/akbasic/sink.h` must still appear in that header, comments and whitespace ignored. Compiling it would only redefine the type, so a compile check could not have found the stale struct, and did not. Registered as the CTest case docs_examples in both configurations. Fixing the four wrong examples turned up two interpreter defects, fixed in the previous commit and recorded in TODO.md section 8. MAINTENANCE.md is new: the fence-tag reference, what to do when the case fails, and the conventions that until now only existed inside source comments -- the three test lists and how two of them invert "passed", the sorted verb table, that a golden file is never edited to suit this interpreter, and that a fix gets mutation-checked with a file copy rather than git checkout. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>
2026-07-31 22:41:36 -04:00
```output
CAUGHT Out Of Bounds ON LINE 30
CARRIED ON
```
Two variables are set when the trap fires: **`ER#`** is the error code and **`EL#`** is
the line it happened on. `ERR(ER#)` gives the message text. On a C128 these are called
`ER` and `EL` with no suffix; this dialect has no bare variable names.
**[Chapter 15](15-error-codes.md) is the complete list** of what `ER#` can hold and
what each code means. The short version is that the interpreter's own codes are 512 to
519 and are the only numbers worth comparing against; two codes render as the same
`ERR()` text, so compare the number rather than the text.
`RESUME` comes in three forms:
| Form | Where it goes |
|---|---|
| `RESUME` | back to the line that failed, and tries again |
| `RESUME NEXT` | on to the line after the one that failed |
| `RESUME (line)` | to a line you name |
Bare `RESUME` only terminates if the handler fixed whatever was wrong — that is what
it is for.
`TRAP` with no argument turns trapping off. An error *inside* a handler is reported
normally rather than re-entering it, so a broken handler cannot loop forever.
## STOP, END and CONT
`STOP` stops the program and returns to the prompt; `CONT` resumes from there. `END`
stops it without arming `CONT`. `QUIT` ends the interpreter itself.