- Sep 07, 2017
-
-
anonym authored
-
anonym authored
-
anonym authored
-
anonym authored
-
anonym authored
-
127.0.0.1 authored
-
anonym authored
-
anonym authored
If the script fails at various points, these files would linger.
-
anonym authored
Comparing the new temporary POT files we generate with the temporary POT files we generated last time (if ever, and if we did, for which branch?) is not relevant; these POT files are only used for merging into a new tails.pot and *that* one is relevant to diff against the old tails.pot.
-
anonym authored
-
anonym authored
-
anonym authored
Merge branches 'bugfix/13439-reproducible-desktop-file-utils-mimeinfo.cache', 'bugfix/13442-reproducible-gdk-pixbuf-loaders.cache', 'bugfix/13441-reproducible-glib2.0-giomodule.cache' and 'bugfix/13440-reproducible-gtk2.0-immodules.cache' into feature/3.2-alpha1
-
anonym authored
-
anonym authored
-
anonym authored
Will-fix: #13439-reproducible-desktop-file-utilsmimeinfo.cache
-
anonym authored
Will-fix: #13442-reproducible-gdk-pixbufloaders.cache
-
anonym authored
Will-fix: #13441-reproducible-glib2.0giomodule.cache
-
anonym authored
Will-fix: #13440-reproducible-gtk2.0immodules.cache
-
intrigeri authored
-
intrigeri authored
-
intrigeri authored
-
intrigeri authored
-
anonym authored
And also make the diff_* functions behave consistently and more like the real diff by printing the diff (not that the printing is what makes the refactoring possible).
-
anonym authored
It would return false if the diff was empty. Let's just forget about the line counting approach...
-
127.0.0.1 authored
No commit message
-
- Sep 06, 2017
-
-
sajolida authored
-
sajolida authored
-
sajolida authored
-
127.0.0.1 authored
-
sajolida authored
-
anonym authored
-
anonym authored
-
anonym authored
Previously we did: if diff_without_pot_creation_date ... && \ diff_pot_only_line_comment_change ... ; then ... but this won't work: the only time both can be true at the same time is if both diffs are truly empty. E.g. if the header date is different, it implies that there is a difference that is not a comment, so the second operand will fail. We have to check for both of these properties at the same time, not separately in isolation.
-
anonym authored
-
anonym authored
Otherwise we'll often converge to a state where we always get into the "Real changes ..." case, which feels wrong; after a single run, consecutive runs should always end up in the "Only header ...# case.
-
sajolida authored
-
sajolida authored
We already have them in Redmine...
-
anonym authored
Counting lines will lose the relationship between additions and additions, e.g. with the old way we wouldn't detect non-comment *additions* to the new file.
-