main contains the wsgibase app. compileapp defined run_models, run_controllers, run_views (called by wsgibase) and they run using restricted (in restricted.py) which does the exec in an environment and captures errors into a RestrictedException object.
Everything else is pretty much decoupled. sqlhtml.py depends on dal.py and html.py. shell.py depends on everything because acts like main but from shell. Massimo On Aug 3, 5:53 pm, Phyo Arkar <phyo.arkarl...@gmail.com> wrote: > They are most core of the webp2y , i c. > On Thu, Aug 4, 2011 at 5:16 AM, Massimo Di Pierro < > > > > > > > > massimo.dipie...@gmail.com> wrote: > > most files in gluon/dal have no dependences. Exception is the triad: > > main.py, compileapp.py, restricted,py. > > > On Aug 3, 5:31 pm, Phyo Arkar <phyo.arkarl...@gmail.com> wrote: > > > Actually more Modularizing inside Gluon should be made easy. > > > > putting all core features into gluon/core , everything that related into > > > gluon/dal , those can be removed without problem into its own folders > > etc. > > > > I will look into this first. > > > > On Thu, Aug 4, 2011 at 4:58 AM, Phyo Arkar <phyo.arkarl...@gmail.com> > > wrote: > > > > > Plus Bandwidth concern. I use my own Dedicated and VPS servers to host. > > > >> They are already busy , and Bigger file-size = more bandwidth , 10MB * > > > >> Download already 10 GB . If this become popular , bandwidth cost will > > be > > > >> great.. > > > > > Errata: > > > > > 10MB * 1000 Downloads already 10GB. My bandwidth limit is 80 GB amonth. > > > > > On Thu, Aug 4, 2011 at 4:55 AM, Phyo Arkar <phyo.arkarl...@gmail.com > > >wrote: > > > > >> Syst admin like make system hardening... > > > > >>> That could be an answer... > > > > >> Yes i am more into "Sys Admin" than web-developer , you've read my > > mind Richard > > > >> Vézina > > > > >> Ok my concern is not really about the File size but the Code Base > > size > > > >> (lines of code , number of features). > > > > >> Here is what i have: > > > > >> 1 . Its best to keep as light ( codebase wise ) as possible . > > > >> 2 . When you know features in this module is not going to be use , > > you > > > >> should be able to remove it. > > > >> 3 . Smaller codebase == Easier to debug > > > >> 4 . Easier to Debug == Easier to Modify == More incentives for > > > >> Contributions > > > >> 5 . Sys admins hates Bloat-ness (even tho we are FAT :P ) > > > >> 6 . Security , easier to control when unwanted modules can be removed. > > > > >> I also still care about file-size too : > > > > >> 14 MB is small for "you" Dosen't mean thats small for other part of > > of > > > >> the world. Here connection speed is Averaged to 64 kbit/s to 256kbit/s > > and > > > >> random (Frequent) disconnects , people believes 1 MB is already big. > > zipped > > > >> web2py is already 6.9MB. > > > > >> Plus Bandwidth concern. I use my own Dedicated and VPS servers to > > host. > > > >> They are already busy , and Bigger file-size = more bandwidth , 10MB * > > > >> Download already 10 GB . If this become popular , bandwidth cost will > > be > > > >> great.. > > > > >> On Wed, Aug 3, 2011 at 9:04 PM, cjrh <caleb.hatti...@gmail.com> > > wrote: > > > > >>> On Wednesday, 3 August 2011 16:22:46 UTC+2, mcm wrote: > > > > >>>> I agree, that size is small even on "embedded" systems by now... > > > >>>> Anyway trying to keep it small is always a good thing IMHO. > > > > >>> I think the word "small" is the wrong word to use here. It doesn't > > > >>> really mean anything. It is a relative word that only has meaning in > > > >>> comparison to something else. If there are files within web2py that > > are > > > >>> not used, then they should be removed regardless of the effect on > > size. I > > > >>> think the OP was saying or suggesting that it should be possible to > > remove > > > >>> files that are not needed on a per-installation basis. But my > > follow-up > > > >>> question to that would be why? Sheer size doesn't seem to be a > > > >>> compelling-enough reason by itself.
