goto-cc: newlines in command files are just whitespace

[tautschnig]

GCC's doc says that options are separated by whitespace, with no
particular significance of newlines. Therefore, process the entire file
at once rather than line-by-line.

Fixes: #6592

• Each commit message has a non-empty body, explaining why the change was made.
• n/a Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
• n/a The feature or user visible behaviour I have added or modified has been documented in the User Guide in doc/cprover-manual/
• Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
• n/a My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[chrisr-diffblue]

Approving because this is an improvement on the existing behaviour, and fixes a known bug. I've left a comment for consideration, but that may just be something to raise as a new bug, rather than to be fixed in this PR.

[chrisr-diffblue]

So this will effectively be replacing newlines with single spaces... it may be an edge case that we really don't care about too much, but just to check have you considered how this may/may not play with the quoting and escaping rules for @files ? It looks to me like theoretically a "backslash, newline" sequence is valid and should be preserved as a newline, for instance.

[tautschnig]

That's a good call-out, but I'm not entirely sure how where I might be able to put such a newline. The following won't work with GCC:

-o
file.gb
-Dhello="  \
"
test.c

With the above, hello just expands to an empty token. I also tried

-o
file.gb
'-Dhello="  \
"'
test.c

but then I even get a warning:

<command-line>:0:7: warning: missing terminating " character

Any ideas what else I might try?

[chrisr-diffblue]

Thats good enough for me to think that "theoretically" doesn't map to "actually supported by gcc in reality" :-)

[codecov]

Codecov Report

Merging #6595 (d9463a1) into develop (6776f0c) will increase coverage by 0.26%.
The diff coverage is 93.75%.

@@             Coverage Diff             @@
##           develop    #6595      +/-   ##
===========================================
+ Coverage    76.21%   76.48%   +0.26%     
===========================================
  Files         1578     1578              
  Lines       181393   181385       -8     
===========================================
+ Hits        138253   138736     +483     
+ Misses       43140    42649     -491     

Impacted Files	Coverage Δ
unit/util/irep.cpp	99.48% <80.00%> (-0.52%)	⬇️
src/ansi-c/c_typecast.cpp	80.43% <100.00%> (-0.06%)	⬇️
src/ansi-c/c_typecheck_expr.cpp	75.02% <100.00%> (-0.02%)	⬇️
src/cpp/parse.cpp	68.56% <100.00%> (ø)
src/goto-cc/gcc_cmdline.cpp	75.38% <100.00%> (+0.58%)	⬆️
unit/interpreter/interpreter.cpp	100.00% <100.00%> (ø)
...olvers/flattening/c_bit_field_replacement_type.cpp	46.66% <0.00%> (-6.67%)	⬇️
src/goto-programs/xml_expr.cpp	53.98% <0.00%> (-0.29%)	⬇️
src/goto-programs/json_expr.cpp	62.20% <0.00%> (-0.07%)	⬇️
src/ansi-c/padding.cpp	93.46% <0.00%> (-0.03%)	⬇️
... and 8 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ed7a0d3...d9463a1. Read the comment docs.