Re: [fossil-users] Smudge and clean equivalents?
On Sun, Jun 11, 2017 at 3:48 PM, wrote: > > Date: Sun, 11 Jun 2017 20:02:08 +0100 > From: John Pateman > To: Fossil SCM user's discussion > Subject: Re: [fossil-users] fossil-users Digest, Vol 113, Issue 18 > Message-ID: > Content-Type: text/plain; charset="utf-8" > > I agree that choices made in the design of Kicad are unhelpful in this > regard but the format is unlikely to change anytime soon. There is a good > reason for one of the issues - components need annotation to generate a net > list and it is acceptable to change the annotation between iterations. > There is some book keeping between schematic circuit diagram and the pcb > layout which are two separate files and programs. There are also some > changes made when libraries are added but this is managed elsewhere by > caching component library changes. Anyhow, the details of exactly what the > exact issues are and why these design choices have been made is not what is > really important - it is the principle of how to cope with the consequences > that concerns me. > I only meant where the only changes are "convenience" changes. Changing annotation is a change to the design. > I have investigated whether this is likely to change in Kicad and I think > the answer, whilst there is a workable solution in git, is going to be no. Many corporations do not use git or any of the other VCSs commonly used by open source projects. Cannot assume any of those "corporate" VCSs have any particular feature of git. (Though, I admit, most corporations aren't going to use Kicad (for a variety of legal reasons).) > But I really don't want to be forced into using git simply on this issue - > I very much like fossil and it has many more advantages for my workflow. > > In fact, all I need is the 'clean' functionality. In git it is achieved by > defining a filter for a particular file type (in my case a .sch) and then a > filter entry for > git.clean "sed -E 's/#(PWR|FLG)[0-9]+/#\1?/'" > What about "fossil clean" ? ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Smudge and clean equivalents?
Yes, I am unconcerned about the size - the date stamp changes on every ‘change’ to the schematic - even if there is actually no material change to the design. Adding an incidental file to the project, or even opening the project file is considered a ‘change’ and tracked by a change to the date and another possible conflict to resolve when nothing at all apart from the date has changed. The date stamp is actually less important - when you annotate the components an annotation is ascribed to the power input to the circuit. If subsequent changes are made to the schematic, it is better that the power input to the network connection is recalculated. By rewriting the POWER flag to ‘unknown’ avoids these problems. BW JP > On 11 Jun 2017, at 20:36, Richard Hipp wrote: > > On 6/11/17, John Pateman wrote: >> Due to some internal book >> keeping there are changes made to the first line of the file which records a >> date stamp when the file is opened even if nothing else changes. > > Why is it a problem for the VCS to record these changing date stamps? > The change is small and will take up very little space in the > repository, due to delta compression. > > -- > D. Richard Hipp > d...@sqlite.org > ___ > fossil-users mailing list > fossil-users@lists.fossil-scm.org > http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users signature.asc Description: Message signed with OpenPGP ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Smudge and clean equivalents?
On 6/11/17, John Pateman wrote: > Due to some internal book > keeping there are changes made to the first line of the file which records a > date stamp when the file is opened even if nothing else changes. Why is it a problem for the VCS to record these changing date stamps? The change is small and will take up very little space in the repository, due to delta compression. -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Smudge and clean equivalents?
Is there any way to implement functionality equivalent to git smudge and clean in fossil? I would like to use fossil for a number of Kicad electronic projects. Kicad uses simple text files for describing the schematic and circuit layout so is a very suitable candidate for version control. Several users have used git to facilitate this but I would like to use fossil. Due to some internal book keeping there are changes made to the first line of the file which records a date stamp when the file is opened even if nothing else changes. I do not want to reversion every time I open a file! There are a couple of other changes I need but I won't complicate it - if I can understand how I might address this in principle I can see if I can work out an actual workflow. Using git it is possible to parse specific files and ignore very selected changes to it using a filter. (Git clean and a Sed incantation). I also need a way to remove some selective changes to a particular line in another file - a trick that is possible using git smudge before staging them. (Yes, I know fossil doesn't do staging like git) Is there any way I can achieve something similar in fossil? For a more detailed description of what I would like to achieve, the workflow I would like to emulate is described using git on the following page http://jnavila.github.io/plotkicadsch/ JP Sent from my iPhone ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
