Make test_symbols.py compare files line-by-line

[ijan1]

Summary

We currently feed full files to Python's unified_diff.
It's not quite what we want though -
line-by-line comparison makes more sense
(we want to be able to identify missing/unnecessary lines)
and is also easier to parse for humans.
This patch makes sure that we compare one line at a time.

This change pretties up the output formatting in the script.
Output before:

!DEF:/m/s/xINTENT(IN)(Implicit)ObjectEntityREAL(4)
!DEF:/m/s/yINTENT(INOUT)(Implicit)ObjectEntityREAL(4)
-!-D-E-F-:-f-o-o-b-a-r-
puresubroutines(x,y)bind(c)
!REF:/m/s/x
intent(in)::x

Proposed output after:

!DEF:/m/s/xINTENT(IN)(Implicit)ObjectEntityREAL(4)
!DEF:/m/s/yINTENT(INOUT)(Implicit)ObjectEntityREAL(4)
-!DEF:foobar
puresubroutines(x,y)bind(c)
!REF:/m/s/x
intent(in)::x

[awarzynski]

Makes a lot of sense, but I would rephrase the summary a bit. Basically, you are making sure here that you are comparing the files line by line rather than the whole thing at a time. By making sure that the comparison is done line-by-line, you also make the output much nicer. Perhaps:

Make test_symbols.py compare files line-by-line

We currently feed full files to Python's unified_diff. It's not quite what we want though - line-by-line comparison makes more sense (we want to be able to identify missing/unnecessary lines) and is also easier to parse for humans. This patch makes sure that we compare one line at a time.

This change pretties up the output formatting in the following way:
Output before:

--- after
+++ before
@@ -6,13 +6,13 @@
!DEF:/m/s/xINTENT(IN)(Implicit)ObjectEntityREAL(4)
!DEF:/m/s/yINTENT(INOUT)(Implicit)ObjectEntityREAL(4)
-!-D-E-F-:-f-o-o-b-a-r-
puresubroutines(x,y)bind(c)
!REF:/m/s/x
intent(in)::x

Proposed output after:

--- after
+++ before
@@ -6,13 +6,13 @@
!DEF:/m/s/xINTENT(IN)(Implicit)ObjectEntityREAL(4)
!DEF:/m/s/yINTENT(INOUT)(Implicit)ObjectEntityREAL(4)
-!DEF:foobar
puresubroutines(x,y)bind(c)
!REF:/m/s/x
intent(in)::x

What do you think? (btw, line numbers ^^^ need tweaking)

[Meinersbur]

flang/test/Semantics/test_symbols.py

46–48

[ijan1]

Updated to use join with unified_diff.

[ijan1]

In D107954#2941348, @awarzynski wrote:

Makes a lot of sense, but I would rephrase the summary a bit. Basically, you are making sure here that you are comparing the files line by line rather than the whole thing at a time. By making sure that the comparison is done line-by-line, you also make the output much nicer. Perhaps:

Thanks for the suggested summary! I've updated accordingly.

[awarzynski]

LGTM, thanks!

[ijan1]

@Meinersbur is this ready to be merged?

[Meinersbur]

yes

flang/test/Semantics/test_symbols.py

 #!/usr/bin/env python3
 
 """Compiles a source file with "-fdebug-unparse-with-symbols' and verifies
 we get the right symbols in the output, i.e. the output should be
 the same as the input, except for the copyright comment.
 Expects a source file passed as the first argument;
-Expects the Flang frontdriver with options as second argument."""
+Expects the Flang frontend driver with options as second argument."""
 
 import sys
 import re
 import subprocess
 import common as cm
 
 from difflib import unified_diff
 

 cmd = [flang_fc1, *flang_fc1_args, flang_fc1_options]
 diff3 = subprocess.check_output(cmd, input=diff2, universal_newlines=True)
 
 # Removes all whitespace to compare differences in files
 diff1 = diff1.replace(" ", "")
 diff3 = diff3.replace(" ", "")
 diff_check = ""
 
 # Compares the input with the output
-for line in unified_diff(diff1, diff3, n=999999,
-                         fromfile="Expected output", tofile="Actual output"):
-    diff_check += line
+diff_check = "\n".join(unified_diff(diff1.split("\n"), diff3.split("\n"), n=999999,
+                       fromfile="Expected_output", tofile="Actual_output"))
 
 if diff_check != "":
     print(diff_check.replace(" ", ""))
     print()
     print("FAIL")
     sys.exit(1)
 else:
     print()
     print("PASS")