On Fri, Jul 03, 2009 at 10:50:17PM +0200, Gianluca Montecchi wrote: > > Hello to everyone > > As i said in a previous mail, I am working on a "html" command for be. > The goal is to be able to do something like "be html /web/page" to have in the > /web/page directory some static html pages that basically are the dump of the > be repository, much like ditz have > This will enable a simple and fast publish of the bus list and details on the > web, at least in read only mode. > > So I'd like to ask some question: > 1) is it ok to develop this command ? I know that this is not a fully featured > web interface, but I am sure that it can be usefull. > > I am open to suggestion about it of course. > > 2) I see that every command is implemented with a python file in the becommand > dir. For a better code, I'd like to split the command implementation into two > files: a file that contain the actual code and a second file that have the html > related part, any problem with this ? I don't like to have the html part and > the code part in one big and unreadable file. > > I'd like to hear other opinion about this. > > Thanks for now > bye > Gianluca > > > _______________________________________________ > Be-devel mailing list > Be-devel@bugseverywhere.org > http://void.printf.net/cgi-bin/mailman/listinfo/be-devel On Mon, Jul 06, 2009 at 10:18:33PM +0200, Gianluca Montecchi wrote: > This sound like an interesting idea, but what i'd like to do is not, strictly > speaking, a report. It is a full tree of html pages that are browseable, both > on line and offline I'm not sure what distinction you're making about "report". You're just producing a static snapshot of the current database status, right? The number of pages and completeness of coverage are nice, but it's still a static entity generated from a particular snapshot, which is what I mean by "report" ;). > > > 2) I see that every command is implemented with a python file in the > > > becommand dir. For a better code, I'd like to split the command > > > implementation into two files: a file that contain the actual code and > > > a second file that have the html related part, any problem with this ? > > > > This sounds quite sensible to me. The existence of a command implies a > > module of the same name in ‘becommand’, but there's no necessary > > implication that that module can't import modules from elsewhere to do > > its work. > > The "elsewhere" for now is the same directory, just another module > On Mon, Jul 06, 2009 at 10:38:56PM +0200, Gianluca Montecchi wrote: > > On Fri, Jul 03, 2009 at 10:50:17PM +0200, Gianluca Montecchi wrote: > > > The goal is to be able to do something like "be html /web/page" to have > > > in the /web/page directory some static html pages that basically are the > > > dump of the be repository, much like ditz have > > > > I think truly interactive frontends like Steve's working on need to be > > build on top of libbe directly, since they'll need to make lots of > > small changes to the database, and it's to slow to be reloading the > > database for every change. Static dumps like my mbox or Gianluca's > > html could just parse the xml output of `be list' and other be > > commands. > > Ok, but if I want to have an html dump that is browseable, I need to parse the > xml. Am I correct ? > If yes, should not be easiear to use directly the libbe ? Using libbe directly is easier, but also more tightly tied to the be internals which could weigh down future refactoring. Partly I'm afraid of our 2.5 different html-output mechanisms. Either their should be a single Right Way that tries to satisfy everyone, or a smorgasbord of loosely coupled translators, so it's not so painful to kill them if/when they go out of style :p. On Mon, Jul 06, 2009 at 10:46:54PM +0200, Gianluca Montecchi wrote: > On Saturday 04 July 2009 02:31:26 Chris Ball wrote: > > It might be a good idea for "be html" to use the CherryPy web interface > > that Steve is working on. The command could start up the CherryPy app > > and scrape all of the available pages to get a stand-alone dump; this > > would avoid having to keep two (okay, more than two at this point) > > separate sets of HTML templates in the source tree. What do you think? > > It can be do, but this implies that CherryPy must be installed and configured, > a thing that I don't want to impose. My idea is to offer a simpler way to have > some html pages, where you just need to have BE installed. I agree that not needing CherryPy for a static html dump is good. Also, read-only templates will look different from the CherryPy interactive templates. +1 for another quasi-redundant template set ;). > > > 2) I see that every command is implemented with a python file in > > > the becommand dir. For a better code, I'd like to split the > > > command implementation into two files: a file that contain the > > > actual code and a second file that have the html related part, > > > any problem with this ? I don't like to have the html part and > > > the code part in one big and unreadable file. > > > > I agree that becommands/*.py commands should not contain any HTML > > layout code. Putting it somewhere else instead sounds fine. > > I am in doubt with the "somewhere else", since for now I put the html template > into a separate file in the same directory. Suggestion ? I think that only code intended only for command line use only should go into becommands, but really, just dump it anywhere and we can shift it around later :p. -- This email may be signed or encrypted with GPG (http://www.gnupg.org). The GPG signature (if present) will be attached as 'signature.asc'. For more information, see http://en.wikipedia.org/wiki/Pretty_Good_Privacy My public key is at http://www.physics.drexel.edu/~wking/pubkey.txt