LRF output - Page 43

JSWolf · 03-05-2008, 01:51 PM

Yeah, it does support bold & Italic.The same way.. simulating it.

curtw · 03-05-2008, 06:16 PM

Quote:

Originally Posted by kovidgoyal

The reader has neither bold nor italic typefaces. It simulates both by hacking the fonts (it thickens the font for bold and slants it for italics). I dont remember if it supports both bold and italic at the same time, one way to test is to see the file in libprs500's inbuilt viewer.

The libprs500 built-in viewer didn't capture the sans-serif part (but the Reader did), so I'm not using it as a yardstick for the bold+italic combo. I'm guessing that the "italic" just aren't slanted *enough* to letme notice a difference.

beowulf573 · 03-05-2008, 08:09 PM

Has anyone experienced a long scan then reboot issue when using the latest (0.4.40) html2lrf? I have a html file, no external links or images, that works fine using version 0.3.106. If I build the lrf file using 0.4.40 and copy it to my SD card using the command line, then try to load it in the reader, the reader begins to scan and format the document, pauses for quite a while, and then reboots.

Other html files using similar formatting seem to work fine.

I'll try to find time tomorrow during lunch to narrow down what's causing the problem in the html file, probably by cutting out chunks until it works. :-)

kovidgoyal · 03-05-2008, 10:44 PM

There were quite a few changes to the html2lrf engine between 0.3 and 0.4.40, some of them allow html2lrf to support more advanced features at the cost of a higher complexity of output LRF. If you can narrow down exactly what in the HTML file is causing the LRF renderer to crash, that would be a big help.

JSWolf · 03-05-2008, 11:09 PM

Quote:

Originally Posted by curtw

The libprs500 built-in viewer didn't capture the sans-serif part (but the Reader did), so I'm not using it as a yardstick for the bold+italic combo. I'm guessing that the "italic" just aren't slanted *enough* to letme notice a difference.

Yes the italic are slanted enough to notice the difference.

JSWolf · 03-05-2008, 11:13 PM

I did find a bug in html2lrf. I had it try to create an LRF from an eBook with code that told it to make the entire book a table at 70% width. I got a resulting LRF that was huge and it caused lrfviewr to not work. Also, it locked up eBook Library's viewer as well. When I edited the HTML to remove this table code, it worked fine as far as the LRF goes. If you need, I can post a link to the HTML that caused this problem. It's one of the eBooks I've recently converted and posted here.

kovidgoyal · 03-05-2008, 11:43 PM

Nah tables are a problem. Just use the --ignore-tables switch. I can't do anything about that because LRF has no real table support and the support html2lrf is kinda hackish.

JSWolf · 03-05-2008, 11:45 PM

Actually, ignoring tables would not have worked as there are tables within that table that I needed to keep as a table.

One thing I sometimes do for a table is create it as a graphic and I'm sure then it'll work.

beowulf573 · 03-06-2008, 12:05 PM

I was able to narrow down the lockup to one css tag, I created a ticket with a sample file but for everyone else here's an example.

This attribute, "margin-left: 25.5em;", seems to be the culprit.

Code:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
<html>
<head>
<title>Sample</title>
<style type="text/css">
 body
 {
     margin-left: 0em;
     margin-right: 0em;
 }
 .attribution1
 {
     font-family: "Times New Roman",serif;
     margin-top: 0.5em;
     margin-bottom: 1.1em;
     margin-left: 25.5em;
 }
 </style>
</head>
<body>

<p class="attribution1">
sample text that doesn't seem to work
</p>
</body>
</html>

bkilian · 03-09-2008, 04:06 AM

Hey Kovid, I've been trying to use --page-break-before-tag, (with any2lrf), but the thing just appears to completely ignore anything I put there. No matter what I use, it seems to use h[12].

I've been trying to get it working on h[1234], for a book I'm converting. I tried using --force-page-break-before, and that works, but gives me too many page breaks that I don't want (where the tags are close together). There's 29 pages between the two tags I want to page-break, so I don't think it's that they are too close together.

Is there anything special I need to do to get it working?

bkilian · 03-09-2008, 04:24 AM

Quote:

Originally Posted by bkilian

Hey Kovid, I've been trying to use --page-break-before-tag, (with any2lrf), but the thing just appears to completely ignore anything I put there. No matter what I use, it seems to use h[12].

I've been trying to get it working on h[1234], for a book I'm converting. I tried using --force-page-break-before, and that works, but gives me too many page breaks that I don't want (where the tags are close together). There's 29 pages between the two tags I want to page-break, so I don't think it's that they are too close together.

Is there anything special I need to do to get it working?

Ok, with "verbose" on, it seems that my setting does catch one more page break (a h3), but none of my h4 chapter headings. How does it determine how little is too little to break?

kovidgoyal · 03-09-2008, 02:14 PM

As far as I remember, it goes by the number of paragraphs, more than 3 means break. Less means ignore. Something like that.

bkilian · 03-09-2008, 04:28 PM

Quote:

Originally Posted by kovidgoyal

As far as I remember, it goes by the number of paragraphs, more than 3 means break. Less means ignore. Something like that.

Then I'm afraid it's broken somewhere.
I got mine working by modifying the chapter detection regex, but without that it never broke on any of my h4's even though they were significantly more than 3 paragraphs apart.

kovidgoyal · 03-09-2008, 04:48 PM

Hmm I've made a change that should fix it.

balok · 03-09-2008, 07:18 PM

Kovid, I've been trying to set up a custom profile for a newspaper, but I can't seem to get the encoding right. The curly apostrophes (u+2019) don't show up in the lrf file on the reader (they show up in your lrf viewer, but not on the prs-500).

Here's a sample page from the newspaper's site:
http://www.cyberpresse.ca/apps/pbcs....intart&print=1

And here's the custom profile:
https://libprs500.kovidgoyal.net/att...cyberpresse.py

03-05-2008, 08:09 PM	#633
beowulf573 Addict Posts: 208 Karma: 1523 Join Date: Jul 2007 Location: Houston,TX Device: PRS-T1	generated lrf file causing lockup? Has anyone experienced a long scan then reboot issue when using the latest (0.4.40) html2lrf? I have a html file, no external links or images, that works fine using version 0.3.106. If I build the lrf file using 0.4.40 and copy it to my SD card using the command line, then try to load it in the reader, the reader begins to scan and format the document, pauses for quite a while, and then reboots. Other html files using similar formatting seem to work fine. I'll try to find time tomorrow during lunch to narrow down what's causing the problem in the html file, probably by cutting out chunks until it works. :-)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Change font of header for LRF Output on PRS 505	duckbill	Calibre	3	05-15-2010 11:07 AM
Pissed off with LRF formatting: LRF/LRS clean tool?	grimborg	LRF	8	02-15-2010 01:14 PM
Fonts for LRF output	krischik	Calibre	1	10-03-2009 05:01 AM
CBZ > LRF (LRF>HTML/MOBI????)	sideburnt	Calibre	4	09-15-2009 06:44 AM
libprs500 Issues Converting .LIT to .LRF - .LRF crashes everything	vasbinde	Calibre	6	02-14-2008 12:16 PM

03-05-2008, 01:51 PM	#631
JSWolf Resident Curmudgeon Posts: 79,758 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	Yeah, it does support bold & Italic.The same way.. simulating it.

03-05-2008, 10:44 PM	#634
kovidgoyal creator of calibre Posts: 45,356 Karma: 27182818 Join Date: Oct 2006 Location: Mumbai, India Device: Various	There were quite a few changes to the html2lrf engine between 0.3 and 0.4.40, some of them allow html2lrf to support more advanced features at the cost of a higher complexity of output LRF. If you can narrow down exactly what in the HTML file is causing the LRF renderer to crash, that would be a big help.

03-05-2008, 11:13 PM	#636
JSWolf Resident Curmudgeon Posts: 79,758 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	I did find a bug in html2lrf. I had it try to create an LRF from an eBook with code that told it to make the entire book a table at 70% width. I got a resulting LRF that was huge and it caused lrfviewr to not work. Also, it locked up eBook Library's viewer as well. When I edited the HTML to remove this table code, it worked fine as far as the LRF goes. If you need, I can post a link to the HTML that caused this problem. It's one of the eBooks I've recently converted and posted here.

03-05-2008, 11:43 PM	#637
kovidgoyal creator of calibre Posts: 45,356 Karma: 27182818 Join Date: Oct 2006 Location: Mumbai, India Device: Various	Nah tables are a problem. Just use the --ignore-tables switch. I can't do anything about that because LRF has no real table support and the support html2lrf is kinda hackish.

03-05-2008, 11:45 PM	#638
JSWolf Resident Curmudgeon Posts: 79,758 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	Actually, ignoring tables would not have worked as there are tables within that table that I needed to keep as a table. One thing I sometimes do for a table is create it as a graphic and I'm sure then it'll work.

03-09-2008, 04:06 AM	#640
bkilian Zealot Posts: 131 Karma: 24870 Join Date: Oct 2006 Device: Sony PRS/505	Hey Kovid, I've been trying to use --page-break-before-tag, (with any2lrf), but the thing just appears to completely ignore anything I put there. No matter what I use, it seems to use h[12]. I've been trying to get it working on h[1234], for a book I'm converting. I tried using --force-page-break-before, and that works, but gives me too many page breaks that I don't want (where the tags are close together). There's 29 pages between the two tags I want to page-break, so I don't think it's that they are too close together. Is there anything special I need to do to get it working?

03-09-2008, 02:14 PM	#642
kovidgoyal creator of calibre Posts: 45,356 Karma: 27182818 Join Date: Oct 2006 Location: Mumbai, India Device: Various	As far as I remember, it goes by the number of paragraphs, more than 3 means break. Less means ignore. Something like that.

03-09-2008, 04:48 PM	#644
kovidgoyal creator of calibre Posts: 45,356 Karma: 27182818 Join Date: Oct 2006 Location: Mumbai, India Device: Various	Hmm I've made a change that should fix it.

03-09-2008, 07:18 PM	#645
balok Ugly alien Posts: 144 Karma: 225 Join Date: Sep 2007 Location: Québec, QC Device: tricorder	Kovid, I've been trying to set up a custom profile for a newspaper, but I can't seem to get the encoding right. The curly apostrophes (u+2019) don't show up in the lrf file on the reader (they show up in your lrf viewer, but not on the prs-500). Here's a sample page from the newspaper's site: http://www.cyberpresse.ca/apps/pbcs....intart&print=1 And here's the custom profile: https://libprs500.kovidgoyal.net/att...cyberpresse.py