Commit Graph
14 Commits
Author SHA1 Message Date
Nicolas Williams 1a1804afcb Fix LOADV/LOADVN refcount printing 2021-10-24 16:23:26 -05:00
Nicolas Williams 582717a7b4 Fix crash in LOADVN when stack grows
This `stack_push()` call in LOADVN invalidates `var`:

       jv* var = frame_local_var(jq, v, level);
       jv_free(stack_popn(jq));
------>stack_push(jq, *var);
       *var = jv_null();
       ^^^^^^

We have to re-compute `var`:

       jv* var = frame_local_var(jq, v, level);
       jv_free(stack_popn(jq));
       stack_push(jq, *var);
------>var = frame_local_var(jq, v, level);
       *var = jv_null();
2021-10-24 16:23:26 -05:00
Leonid S. Usov cf4b48c7ba Save literal value of the parsed number to preserve it for the output
Extend jv_number to use decNumber for storing number literals. Any math
operations on the numbers will truncate them to double precision.
Comparisons when both numbers are literal numbers will compare them
without truncation.

Delay conversion of numbers to doubles until a math operation is performed,
to preserve precision. A literal jv_number will only need conversion to
double once, and will reuse the resultant double on subsequent
conversions.

Outputting literal jv_numbers preserves the original precision.

Add strong pthread requirement to manage contexts/allocations for
converting numbers between their decNumber, string, and double formats.
2019-10-22 14:11:04 -04:00
Nicolas Williams b52fc1043b Fix assert in generator subexpressions (fix #1875)
Expressions of the form `path(EXPR) | select(GENERATOR)`, where `EXPR`
is a path expression and `GENERATOR` is a generator conditional
expression (e.g., `has("a"), has("b")`) cause an assertion if the
jq_state VM is torn down too soon.  That assert() was only correct if
assuming that the conditional is not a generator.

If the conditional is generator, then what we see is that when
backtracking a SUBEXP_END is executed without a corresponding
SUBEXP_BEGIN because the entire conditional is bracketed with
SUBEXP_BEGIN and SUBEXP_END, and since it's resumed in the middle, in
between the brackets.

Rather than assert that the jq->path_len being restored has some
particular value, we can simply re-compute it from the restored
jq->path.
2019-03-26 18:58:55 -05:00
William Langford dd4cec664c fix memory leak 2018-08-30 21:09:04 -04:00
William Langford 0673dee1b3 Fix destructuring alternation
Attempting to use the existing FORK_OPT opcode resulted in difficulty
knowing when to pop an error message off the stack and when not to. This
commit makes DESTRUCTURE_ALT a real opcode that is identical to
FORK_OPT, except for never pushing the error message onto the stack when
continuing from an error backtrack.

Some small changes were necessary to the DUP/POP behavior surrounding
destructuring to accomodate this.
2018-08-17 23:15:48 -04:00
William Langford 94035be3fa Add --debug-trace=all for detailed debug output
When tracing execution, print the whole stack, not just the items used by the
current opcode
2017-03-26 05:36:22 -05:00
William Langford 3a8c8f4747 Add alternation destructuring operator ?//
This is a first pass to show the implementation.
It needs tests and evaluation, but doesn't break any existing tests.

NOT READY FOR MERGING
2017-03-26 05:36:22 -05:00
Nicolas Williams 2fcb257168 getpath/1 should be a path expression (fix #1358)
It needs to be possible to do something like

    getpath($paths[]) += 1

meaning: increment all the paths in . that are listed in $paths[].

In order to do this getpath() needs to update the jq->path and
jq->value_at_path as necessary.
2017-02-28 21:54:30 -06:00
Nicolas Williams 8ea21a54ad Add halt, halt_error builtins (fix #386) 2017-02-26 16:34:56 -06:00
Nicolas Williams a03ae02f44 Fix memory leak 2017-02-26 16:33:50 -06:00
Nicolas Williams b279713e47 fixup 2017-02-26 16:12:57 -06:00
Nicolas Williams e8abc0a2f6 jq_compile_args(): allow object args to be object 2017-02-25 19:13:28 -06:00
David Tolnay 0c93eb3379 Move source files to src/ 2015-08-23 20:36:11 -07:00