[PATCH 5 of 6 V3] hgweb: blacklist heavyweight revset functions in hgweb search

[Alexander Plavin]

13.09.2013, 21:59, "Kevin Bullock" <kbullock+mercurial at ringworld.org>:
> On 13 Sep 2013, at 3:43 AM, Alexander Plavin wrote:
>
>>  13.09.2013, 07:12, "Kevin Bullock" <kbullock+mercurial at ringworld.org>:
>>>  Alexander, one thing that might be helpful here is if you use a --flag for each series that ties the different _versions_ of a series together. (Clearly that doesn't help for series you've already sent several versions of.)
>>  Do you mean a flag that remains constant for all versions of a series?
>
> Yes, and preferably something terse and descriptive (not just A, B, C or series1, series2).

It's an important addition :) At first I thought about random ids.

>
>>  Will try this, however there may be problems with a part of a series being accepted and another part - not.
>
> I don't think that's a huge problem -- it'll be fairly obvious if the flag stays the same but N in [M of N] decreases (or increases!). We fairly often ask contributors to either split or rearrange patches (as I did with your revset search series).

Should I resend all already sent series with such flags?

>
> pacem in terris / мир / शान्ति / ‎‫سَلاَم‬ / 平和
> Kevin R. Bullock