New HTML parser integrated into NetSurf
Hi, This morning, we merged into NetSurf the HTML parser that Andrew Sidwell has been working on as his Google Summer of Code project. The new parser is significantly more robust than the old one, particularly in the face of invalid input (which is in excess of 90% of the pages on the web). This change resolves more than 40 bug reports, which we will be closing in the near future. John.
Re: New HTML parser integrated into NetSurf
In article <[EMAIL PROTECTED]>, John-Mark Bell <[EMAIL PROTECTED]> wrote: > Hi, > This morning, we merged into NetSurf the HTML parser that Andrew > Sidwell has been working on as his Google Summer of Code project. It doesn't seem to take any notice of the background color in a Table. see:- http://www.nowirul.org.uk/halbro/draw0809/test.htm Should be a dark green border around each cell. -- Barry A.
Re: New HTML parser integrated into NetSurf
In a dim and distant universe <[EMAIL PROTECTED]>, Barry E Allen <[EMAIL PROTECTED]> enlightened us thusly: > It doesn't seem to take any notice of the background color in a Table. > see:- Something has broken WGT to table colouring, as the latest build breaks the listing at http://ccdb.cropcircleresearch.com/?d=x&y=2008&c=uk&m=Aug The only dark green background should be the column headings at the top. All the data rows should alternate between light green and white graph paper. The rows that should be white (same colour as the page background) are coming out in dark green (same colour as the top row titles). Paul Vigay Editor, www.RISCOS.org -- Using, programming and promoting RISC OS - the most productive computer system in the world. Check it out now, and change your view of computers!! To reply/email, visit http://www.riscos.org/feedback/ The two most common elements in the universe are hydrogen and stupidity.
Re: New HTML parser integrated into NetSurf
In message <[EMAIL PROTECTED]> John-Mark Bell <[EMAIL PROTECTED]> wrote: >Hi, > >This morning, we merged into NetSurf the HTML parser that Andrew Sidwell >has been working on as his Google Summer of Code project. The new parser >is significantly more robust than the old one, particularly in the face of >invalid input (which is in excess of 90% of the pages on the web). This >change resolves more than 40 bug reports, which we will be closing in >the near future. > > >John. > Thanks for all the hard work by the team, it also seems to have fixed the UTF8 bad character bug, which when it found a bad UTF8 character it displayed gooblygook. Well done. :) -- Kev Wells http://riscos.kevsoft.co.uk/ http://kevsoft.co.uk/ http://kevsoft.co.uk/AleQuest/ ICQ 238580561 IF your nose runs and your feet smell it means your upside down.
Re: New HTML parser integrated into NetSurf
In message <[EMAIL PROTECTED]> John-Mark Bell <[EMAIL PROTECTED]> wrote: > Hi, > > This morning, we merged into NetSurf the HTML parser that Andrew Sidwell > has been working on as his Google Summer of Code project. The new parser > is significantly more robust than the old one, particularly in the face of > invalid input (which is in excess of 90% of the pages on the web). This > change resolves more than 40 bug reports, which we will be closing in > the near future. Well done and thank you to all the NetSurf team. This one makes Barclaycard statements and transactions visible. Another useful site working! Dave
Maplin site: central section is too wide
Why is it that the central section of the Maplin web site http://www.maplin.co.uk/ is too wide and therefore gets pushed down the page a long way? The discrepancy is just a very few pixels. Dave
Re: Maplin site: central section is too wide
On Mon, 11 Aug 2008, Dave Higton wrote: Why is it that the central section of the Maplin web site http://www.maplin.co.uk/ is too wide and therefore gets pushed down the page a long way? The discrepancy is just a very few pixels. Sounds like a rounding error. Please report this on the bug tracker. Thanks, John.
Re: New HTML parser integrated into NetSurf
On Mon, 11 Aug 2008, Barry E Allen wrote: In article <[EMAIL PROTECTED]>, John-Mark Bell <[EMAIL PROTECTED]> wrote: Hi, This morning, we merged into NetSurf the HTML parser that Andrew Sidwell has been working on as his Google Summer of Code project. It doesn't seem to take any notice of the background color in a Table. see:- http://www.nowirul.org.uk/halbro/draw0809/test.htm Should be a dark green border around each cell. Please report this on the bug tracker, attaching the above page as a testcase. Thanks, John.
Re: New HTML parser integrated into NetSurf
On 11 Aug 2008 Kevin Wells <[EMAIL PROTECTED]> wrote: > Thanks for all the hard work by the team, it also seems to have fixed > the UTF8 bad character bug, which when it found a bad UTF8 character it > displayed gooblygook. I can confirm that, the PocketGPSWorld newsletter would always be truncated half way down, as the HTML has charset=UTF-8, but '£' is used illegally instead of characters A0 A3. I'll be waiting for the table bug to be fixed before adopting the new parser though. Cheers ---Dave -- __ David J. Ruck Phone: +44- (0)7974 108301 Email: [EMAIL PROTECTED] __
Re: Maplin site: central section is too wide
On 11 Aug 2008 John-Mark Bell wrote: > On Mon, 11 Aug 2008, Dave Higton wrote: >> Why is it that the central section of the Maplin web site >> http://www.maplin.co.uk/ is too wide and therefore gets pushed >> down the page a long way? The discrepancy is just a very few >> pixels. > Sounds like a rounding error. Please report this on the bug tracker. I'm pleased that the problem where items intended to be in a vertical column were progressively shifted further and further to the right has been solved. Richard -- _ |_|. _ Richard Porter http://www.minijem.plus.com/ |\_||_mailto:[EMAIL PROTECTED]
Spurious
I've found a very curious problem. At the bottom of a long table which contains several forms I get a spurious "<" character. I have made a change but I can't see a missing ">" or missing quote or anything that would likely to cause it. When I stick in a " at any point between table rows to narrow down the location the spurious character dissapears. It doesn't appear on other browsers. I can't post a bug report because the html file is local and contains hidden logins which I would have to alter. -- _ |_|. _ Richard Porter http://www.minijem.plus.com/ |\_||_mailto:[EMAIL PROTECTED]
Re: Spurious
In message <[EMAIL PROTECTED]> Richard Porter <[EMAIL PROTECTED]> wrote: >I've found a very curious problem. At the bottom of a long table which >contains several forms I get a spurious "<" character. I have made a >change but I can't see a missing ">" or missing quote or anything that >would likely to cause it. When I stick in a " at any >point between table rows to narrow down the location the spurious >character dissapears. It doesn't appear on other browsers. Have you tried doing a search in your text editor for < and > and count the number of time each occurs? StrongED has a list of found option which counts them for you. > >I can't post a bug report because the html file is local and contains >hidden logins which I would have to alter. > -- Kev Wells http://riscos.kevsoft.co.uk/ http://kevsoft.co.uk/ http://kevsoft.co.uk/AleQuest/ ICQ 238580561 On England's pleasant pastures seen?
Re: Spurious
On 11 Aug 2008 Kevin Wells wrote: > In message <[EMAIL PROTECTED]> > Richard Porter <[EMAIL PROTECTED]> wrote: >>I've found a very curious problem. At the bottom of a long table which >>contains several forms I get a spurious "<" character. I have made a >>change but I can't see a missing ">" or missing quote or anything that >>would likely to cause it. When I stick in a " at any >>point between table rows to narrow down the location the spurious >>character dissapears. It doesn't appear on other browsers. > Have you tried doing a search in your text editor for < and > and count > the number of time each occurs? No, I've checked the changes I made very carefully. I think if there was an extra "<" somewhere it would show up all the time. It is definitely not mismatched < and > but I have narrowed down the problem. One of the forms has an 'action' attribute which, including encoded ampersands, quotes, etc. is 257 characters long. If I move the closing ">" onto the next line the problem goes away. Other tweaks caused the following table to be corrupted. I suspect that the new parser is getting phased by the long string. -- _ |_|. _ Richard Porter http://www.minijem.plus.com/ |\_||_mailto:[EMAIL PROTECTED]
