news from the topic experiment

Pierre-Yves David pierre-yves.david at
Fri Sep 16 08:53:11 EDT 2016

On 09/15/2016 07:27 PM, Durham Goode wrote:
> On 9/14/16 12:14 PM, Pierre-Yves David wrote:
>> Example output for messy state:
>>> ### topic: vfs.ward (2 heads)
>>> ### branch: default, 189 behind
>>> t7$ reposvfs: add a ward to check if locks are properly... (unstable)
>>> t6@ mq: release lock after transaction in qrefresh (current)
>>> t5: perf: release lock after transaction in perffncachewrite
>>> t3^ repovfs: add a ward to check if locks are properly taken (base)
>>> t4$ ignore bisect (unstable)
>>> t3: repovfs: add a ward to check if locks are properly taken
>>> t2: vfs: add the possibility to have a "ward" to check vfs usage
>>> t1: pull: grab wlock during pull
>>>   ^ journal: take wlock for writting the 'shared' file
> Why not use the graph ascii art to make the ancestry clearer? You can
> still put the 'tX' shortcut in front of each commit.

Partly because 'it simply has not been done yet', This is not just a 
graphlog call, as we have to build the graph edge by hand (but 
definitely doable).

In practice, it is also not that critical, with color enabled, the 
(base) lines stand out a lot.

In addition, stack focus on a single topic that is expected to be 
linear, so any branching is supposed to be a temporary problematic case. 
So the output is not optimisted too much for that case as bugging the 
user until it is solved is an actual goal here. In addition the base can 
also be outside the topic (in which case, the line do not have a t#) and 
the graphlog will not be very informative.

> I had to sit and stare at this output for a while to figure out what was
> going on (in particular, that t3 was present twice).

The glyph next to the changeset short id are a bit cryptic at first. I'm 
open to idea of how to improve that and in the mean time I expect user 
to adjust in a descent amount of time.


Pierre-Yves David

More information about the Mercurial-devel mailing list