Hello everybody The following emails for some reaseon never made to the list so I am resending them. ---------------- Email 1 ---------------------------------------- James C wrote:
src/table.hh might be a reasonable place to start thinking about the requirements for integrating a new set of tags.
Don't know. I looked at it. Are you sure you meant table.hh? I'm not sure that any new tags would actually be added anyway. If anything I would look at Html_tag_open_script(...) { Html_tag_close_script(...) { as likely candidates for a starting point. I did get binaries for iceweasel that worked. 285 files @ 21M and all I do is click: "accept" Thats it. Close the browser(and the chroot, since i can't get it working otherwise) and start dillo. I don't need javascript again until the lease expires about 20hrs later. Bruce --------------- Email 2 ---------------------------------------- James C wrote
Do you have dtrace, dtruss or strace, to look for open() to see where JavaScript is writing its state? That's my guess about why you'd need a chroot or similar.
I have strace, but the reason I need the chroot is that the library versions for iceweasel don't match the versions on my system and the files are all over the place. Iceweasel will eventually go away. Right now getting the glue between dillo and duktake is beginning to look pretty good but I have run into a puzzle of a different kind. I put a print statement in Html_tag_open_script and Html_tag_close_script and indeed you see "in Html_tag_open_script" and "Html_tag_close_script" every time the script tag is encountered. Now if I load that signon page I get a blank screen and the messages are -> not <- displayed, and the stop Icon stays permanently on. Looking at the code I find the tag <noscript> <meta HTTP-EQUIV="REFRESH" content="0; url=https://guestportal.burnaby.ca/login_noscript.html"> </noscript> If I delete these lines dillo loads the page normally and every thing is fine. Looking in the table in html.cc I find: const TagInfo Tags[] = ... ... /* noscript 1011 */ ... ... blindly replaceing this line with: {"script", B8(111001),'R',2, Html_tag_open_script,NULL,Html_tag_close_script}, and changing "script" to "noscript" on the assumtion that when dillo hits the tag </noscript> it will then just continue until it hits </noscript>, throw away the data and then continue on, which it apparently does. The question is what to do about it. If it is something that is going to be implemented but nobody has got around to it then I would at least put in a: Html_tag_open_noscript() and a Html_tag_close_noscript() so that web pages don't stall when dillo hits that tag. Bruce