Matt Harbison <matt_harbison@yahoo.com> [Sun, 08 Jan 2017 02:43:01 -0500] rev 30730
help: apply the section headings from revsets to templates
Unlike filesets, there are a few distinct headings that are not shared with
revsets. But common names are used where possible.
Matt Harbison <matt_harbison@yahoo.com> [Sun, 08 Jan 2017 02:40:36 -0500] rev 30729
help: apply the section headings from revsets to filesets
This has the nice property of visually breaking up the wall of text. It also
allows specific smaller sections to be called out. For example,
`hg help filesets.predicates` now prints just the predicate section. At the
moment, the revset headings are a superset of the fileset headings, so there is
consistency in how example, predicate and operator help is called out.
The reference to `hg help patterns` was moved to the overview section, so that
it isn't stuck in the examples section.
Jun Wu <quark@fb.com> [Fri, 06 Jan 2017 16:14:52 +0000] rev 30728
chg: check type read from S channel
The previous patch added the check server-side. This patch added it
client-side.
Jun Wu <quark@fb.com> [Fri, 06 Jan 2017 16:12:25 +0000] rev 30727
chgserver: check type passed to S channel
It currently only supports the "system" type. Add an explicit check.
Jun Wu <quark@fb.com> [Fri, 06 Jan 2017 16:11:03 +0000] rev 30726
chg: send type information via S channel (BC)
Previously S channel is only used to send system commands. It will also be
used to send pager commands. So add a type parameter.
This breaks older chg clients. But chg and hg should always come from a
single commit and be packed into a single package. Supporting running
inconsistent versions of chg and hg seems to be unnecessarily complicated
with little benefit. So just make the change and assume people won't use
inconsistent chg with hg.
Valters Vingolds <valters@vingolds.ch> [Sun, 01 Jan 2017 13:16:29 +0100] rev 30725
rebase: fail-fast the pull if working dir is not clean (BC)
Refuse to run 'hg pull --rebase' if there are uncommitted changes:
so that instead of going ahead with fetching changes and then suddenly aborting
the rebase, we can warn user of uncommitted changes (or unclean repo state)
right up front.
In tests, we create a 'histedit' session to verify that also an unfinished
state is detected and handled.
Yuya Nishihara <yuya@tcha.org> [Fri, 06 Jan 2017 22:50:04 +0900] rev 30724
commit: fix unmodified message detection for the "--- >8 ----" magic
We need the raw editortext to be compared with the templatetext.
Yuya Nishihara <yuya@tcha.org> [Fri, 06 Jan 2017 22:44:39 +0900] rev 30723
commit: update test to actually modify template text
We have a check for unmodified commit message (introduced by bec1a579ebc4),
which should be enabled for the "--- >8 ---" magic but currently not.