Matt Harbison <matt_harbison@yahoo.com> [Tue, 02 Jan 2018 21:18:30 -0500] rev 35525
lfs: add a local store method for opening a blob
The has() and read() methods already dynamically switch between the usercache
and local store. This should generally be preferred to directly accessing the
vfs instances outside of the store.
The file is now explicitly opened in binary mode for clarity. (It was also
being opened in binary mode before, but only because vfs.__call__() appends 'b'
if needed when not opening with 'text=True'.)
Pulkit Goyal <7895pulkit@gmail.com> [Fri, 05 Jan 2018 17:23:55 +0530] rev 35524
hgdemandimport: use correct hyperlink to python-bug in comments (
issue5765)
Differential Revision: https://phab.mercurial-scm.org/D1810
Anton Shestakov <av6@dwimlabs.net> [Wed, 03 Jan 2018 10:37:49 +0800] rev 35523
monoblue: adjust font size of graph entries
/shortlog uses 1.1em font for changesets, so it makes sense to use the same
size on /graph too. In practice, 1.2em is 14.784px, and 1.1em is 13.552px.
Anton Shestakov <av6@dwimlabs.net> [Wed, 03 Jan 2018 09:41:01 +0800] rev 35522
hgweb: make .info a block element by default
Using a <div> instead of a <span> with "display: block" makes more sense.
While at it, let's adjust top margin in monoblue so that text in .info doesn't
overlap changeset "tags" elements (they are quite large in monoblue).
Gregory Szorc <gregory.szorc@gmail.com> [Mon, 25 Dec 2017 16:31:14 -0700] rev 35521
revlog: don't use slicing to return parents
This is the only place we use a slice on index entries, which
are currently tuples. In preparation for moving away from tuples,
let's stop using slices so we don't have to implement that support
on the new type.
We also tweak the logic slightly so the exception only catches the
IndexError on the index lookup, not on the index entry lookup. The
old code should never have been buggy. But it was semantically wrong.
Differential Revision: https://phab.mercurial-scm.org/D1764
Kyle Lippincott <spectral@google.com> [Thu, 04 Jan 2018 16:29:03 -0800] rev 35520
tests: switch test from '--config ui.editor=~/foo' to HGEDITOR=~/foo
The test is currently not testing anything, the '--config ui.editor=~/path'
argument is apparently of lower priority than the HGEDITOR that run-tests.py is
setting.
Differential Revision: https://phab.mercurial-scm.org/D1809
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 27 Dec 2017 00:24:53 +0530] rev 35519
revert: support reverting to hidden cset if directaccess config is set
This patch adds support for reverting files to a hidden changeset if
`experimental.directacess=True` is set.
Differential Revision: https://phab.mercurial-scm.org/D1763
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 27 Dec 2017 00:14:00 +0530] rev 35518
update: support updating to hidden cset if directaccess config is set
This patch adds support for updating to a hidden changeset without using
--hidden if `experimental.directacces=True` is set. The update command will
print out a warning when updating to a hidden changeset saying:
`updating to a hidden changeset <hash>`
The warning is also printed when directaccess is not used and --hidden is
passed which I think is good behaviour. Tests are added for the directaccess
part and updated output of tests which have case of updating to hidden cset
using `--hidden`.
Differential Revision: https://phab.mercurial-scm.org/D1762