On Wed, 2011-05-04 at 23:38 +0100, Vincent Sanders wrote:

> > Index: render/html_internal.h
> > ===================================================================
> > --- /dev/null       2009-04-16 19:17:07.000000000 +0100
> > +++ render/html_internal.h  2011-05-04 22:29:08.000000000 +0100
> > @@ -0,0 +1,103 @@
> > +/*
> > + * Copyright 2004 James Bursa <[email protected]>
> 
> really? I think its 2011 and you added this file ;-) maybe its a move tho?

The contents of this file were entirely extracted from render/html.h

> > +
> > +/** Data specific to CONTENT_HTML. */
> > +typedef struct html_content {
[...]
> > +} html_content;
> 
> gah make a decision on before or after doccomments within one struct
> ;-) and are we saying 0 or NULL for empty ponters?

Done, and NULL.

> no doccomments?

Indeed not -- they weren't any before, either. At some point, we should
go through and wholesale move comments for public APIs into the headers
where they belong.

> > Index: image/image.c
> > ===================================================================
> > --- /dev/null       2009-04-16 19:17:07.000000000 +0100
> > +++ image/image.c   2011-05-04 22:29:39.000000000 +0100
> 
> 
> > +
> > +nserror image_init(void)
> > +{
> 
> comment? or at least a ref to the header?

Done.

> > +
> > +void image_fini(void)
> 
> ditto

Done.

> > Index: content/content_factory.c
> > ===================================================================
> > --- /dev/null       2009-04-16 19:17:07.000000000 +0100
> > +++ content/content_factory.c       2011-05-04 22:30:17.000000000 +0100
> 
> file comment?

Done.

> > +
> > +typedef struct content_handler_entry {
> 
> 
> > +   struct content_handler_entry *next;
> > +
> > +   lwc_string *mime_type;
> > +   const content_handler *handler;
> > +} content_handler_entry;
> 
> 
> whassat for then? comment teh struct a bit? tbh i shall shut up about
> comments now, its boring

Done.

> > +
> > +static content_handler_entry *content_handlers;
> > +
> > +nserror content_factory_register_handler(lwc_string *mime_type,
> > +           const content_handler *handler)
> > +{
> > +   content_handler_entry *e;
> 
> e? really? perhaps entry might have been a little less brief?

Done.

> > +
> > +content_type content_factory_type_from_mime_type(const char *mime_type)
> > +{
> > +   const content_handler *handler;
> > +   lwc_string *imime_type;
> > +   lwc_error lerror;
> > +   content_type type;
> > +
> > +   lerror = lwc_intern_string(mime_type, strlen(mime_type), &imime_type);
> > +   if (lerror != lwc_error_ok)
> > +           return CONTENT_NONE;
> > +
> > +   handler = content_lookup(imime_type);
> > +   if (handler == NULL) {
> > +           lwc_string_unref(imime_type);
> > +           return CONTENT_NONE;
> > +   }
> > +
> > +   type = handler->type(imime_type);
> > +
> > +   lwc_string_unref(imime_type);
> > +
> > +   return type;
> > +}
> 
> 
> maybe? better as:
> 
>       content_type type = CONTENT_NONE;
> 
>       lerror = lwc_intern_string(mime_type, strlen(mime_type), &imime_type);
>       if (lerror != lwc_error_ok)
>               return CONTENT_NONE;
> 
>       handler = content_lookup(imime_type);
>       if (handler != NULL) {
>               type = handler->type(imime_type);
>         } 
> 
>       lwc_string_unref(imime_type);
> 
>       return type;

Yes, thanks.
  
> >  error:
> >     if (error == BINDING_BADENCODING) {
> > -           LOG(("Bad encoding: %s", html->encoding ? html->encoding : ""));
> > +           LOG(("Bad encoding: %s", c->encoding ? c->encoding : ""));
> >             msg_data.error = messages_get("ParsingFail");
> >     } else
> >             msg_data.error = messages_get("NoMemory");
> >  
> > -   content_broadcast(c, CONTENT_MSG_ERROR, msg_data);
> > -   return false;
> > +   content_broadcast(&c->base, CONTENT_MSG_ERROR, msg_data);
> > +
> > +   return NSERROR_NOMEM;
> 
> always no memory?

Indeed not. Fixed.

> >  
> >  /**
> > @@ -2174,12 +2209,11 @@
> >   */
> >  hlcache_handle *html_get_favicon(hlcache_handle *h)
> 
> kept the API even though the rest of favicon went?

For now, yes -- there are sufficient callsites that removing it is
somewhat invasive (and orthogonal to the point of this changeset)

Ta,


J.


Reply via email to