Firmware 3.18.0 discussion

[JSWolf]

Originally Posted by GeoffR

I think that is intentional. It has been like that forever.

The ePub built-in stylesheet has the following line:

body { padding-bottom: %1 !important; }

where %1 is the height of one line.

In other words, a blank line is added automatically at the end of every chapter, the effect of which is to (sometimes) push the last line onto the next page.

Some publishers also add their own code to do the same thing, sometimes by adding a blank paragraph at the end of a chapter. In some cases this can combine with the stylesheet to push more than one line to the next page.

(Edit: Those who patch the firmware might not see this, depending on their choice of patches, because some of the patches remove this line from the built-in stylesheet to make room for other things.)

Which patches edit out this line from the built-in CSS? I know I've never seen this bug.

However, if I add in body { padding-bottom: 0 !important; } to the book's CSs, would this override the RMDSK CSS?

[doxallday]

Hi everyone.

Spoiler:

I did a factory reset yesterday since I had some troubles with the latest firmware, unfortunately my "Reading Life" is gone (even though I've performed a sync with Kobo prior to going through with it). But that is not the main issue: the reason I did the reset was that during boot-up of my Aura H2O there are multiple loading bars (botched) and once it loads to home screen, it seems to flash once, then I wait few seconds, then it flashes again (which is quite annoying); prior to this patch, I cannot recall having it behave this way.

What I'd like to know, or get assistance with (from you awesome folk), is how do I fix this? After the reset it's all been sunshine and rainbows, but after transferring some books via calibre, powering off, and firing it up again, I get the same distorted loading bars and multiple flashes in home screen (I got about 20 books total at the moment).

Also, can I permanently set margins (via a configuration file on Kobo if possible) that overrides any CSS formatting from loaded books?

If this discussion is not appropriate here, I'm sorry in advance, and will gladly move it elsewhere.

---------
UPDATE: Added the "[FeatureSettingsI" in the .conf file, and deleted all books from my microSD, reformatted the card, and the bars are gone during boot (and it comes back to live quickly as it did before), however, after reboot, I keep getting microSD not recognized and "please format (FAT32)".

The books don't show under Library anymore, however if I click on the tile of the book I read prior to the message (loaded on the microSD) it opens just fine, but when I return to Home Screen, the tile is gone, and there is no way to open it again. What a strange string of events; mildly infuriating. Any clues to why this is happening?

[GeoffR]

Originally Posted by JSWolf

Which patches edit out this line from the built-in CSS? I know I've never seen this bug.

The `ePub fixed/adjustable top/bottom margins` and `Enforce user line spacing and justification in ePubs` patches remove that line.

However, if I add in body { padding-bottom: 0 !important; } to the book's CSs, would this override the RMDSK CSS?

No, the !important in the built-in stylesheet overrides the !important in the book's stylesheet.

[JSWolf]

Originally Posted by GeoffR

The `ePub fixed/adjustable top/bottom margins` and `Enforce user line spacing and justification in ePubs` patches remove that line.

That's odd. I don't use either patch and yet I've never seen this bug. I have text go all the way to the bottom of the screen. Well, right on top of the footer.

[GeoffR]

Originally Posted by JSWolf

That's odd. I don't use either patch and yet I've never seen this bug. I have text go all the way to the bottom of the screen. Well, right on top of the footer.

The last time I saw it was back in firmware 2.8.1 or early 3.0 series I think, which was around two years ago. It could just be a coincidence that it stopped hapenning about the time I started using that patch, a lot of other things have changed since then including the major update of the RMSDK.

Edit: I have experimented with the ePub I am currently reading by adjusting the font size / line spacing / margins to try and make the last line of the chapter end exactly on the last line of the page:

* With the patch to remove the padding-bottom line from the built-in stylesheet, I can make the last line of the chapter end on the last line of the page.

* Without the patch, I can't make the last line of the chapter end on the last line of the page, instead it gets bumped to the next page and I end up with a blank line at the bottom of the (now second-to-last) page.

So I am certain that the padding-bottom in the built-in stylesheet has the effect of pushing the last line of the chapter onto the next page if it was set to appear on the last line of the page, but I'm not sure how it could make a line further up the page get pushed onto the next page unless there is some other interaction with the book's stylesheet (e.g. orphans/widows settings, body margin- or padding-bottom set in the XPGT stylesheet, body font-size set larger than the paragraph font size, etc.)

Have a look in the 3.18.0 patching thread if you want to experiment yourself.

[ploum]

Originally Posted by Nick_1964

I think you ment: it doesn't shut down..
It doesn't when it is connected to the charger, and it doesn't get out always when the casing is closed which is rather strange because with the same casing it always did before. (as is, it doesn't detect the closure right and doesn't went into sleep modus..)
But then,as a fall safe, the timer jumps in and put the device to sleep. (and I had it several times that that didn't happened to, both random..)

Indeed, I mean "it doesn't shut down".

But it seems there's no timer :*I once closed the Kobo and didn't care about it and when I*came back a few hours later, the light was still on.

[Nick_1964]

Originally Posted by ploum

Indeed, I mean "it doesn't shut down".

But it seems there's no timer :*I once closed the Kobo and didn't care about it and when I*came back a few hours later, the light was still on.

In the settings menu there is, translated from my own language something with sleep and energy, there u can make your settings like go to sleep after xx minutes, and shut off after xx and there is also the option to enable the sleepcover.
If you set the sleep modus in 15 minutes or so, even when the sleepcover doesn't put your reader into sleep, the 15 minutes setting will.. ( there are some reports of a bug, in that case, put it to 10 minutes. ) off course, when the flap of the cover is opened, the device wake up.
Do notice that the light went off in 2 quick steps, so between closing the flap and light fully off is about 1,5 seconds..

[Bobbe]

Hello, I'm afraid I have a question that is very possibly pretty stupid.
But I just cannot understand what is happening.

I 'cleaned' and converted epub's to epub's with Calibre. I sent them to my 2 Glo HD devices (which are on firmware 3.16.something and patched) and all was fine. They both showed the 'new' epub's (I told Calibre to delete the empty lines after each paragraph). Fine, no problem so far.

Yesterday I upgraded Aura HD to 3.18.0, and after that I immediately patched it. I sent some of the converted epub's with Calibre to Aura HD. But it shows the epub's with the blank spaces between paragraphs, and not the new cleaned versions. What is happening here? I don't understand...

[Edit]
Found it! It was a setting of the patch that said "Ignore book internal CSS stylesheet", once I set that to 'no' my books were fine. I knew it was probably something stupid like this.

Now my final issue is really the "Full screen reading" issue. I don't like this issue, I like to use the whole screen. I will not update both the Glo HD's, and am thinking about putting an older firmware back on Aura HD as well.

[outlandos]

Well let's hope 3.19.0 is around the corner.

[anguila]

Hello, I would like to ask you fellow book worms if the bugs introduced in this firmware are also present in the Touch versions of the Kobo e-readers?

It seems to me they counter the benefits of the (much wanted) ghosting improvements.

[Shopaholic]

Originally Posted by anguila

It seems to me they counter the benefits of the (much wanted) ghosting improvements.

Welcome to Kobo! That's usually what happens with their updates. It fixes one problem but causes or exacerbates another.

Sadly I cannot answer your question re the Touch because I don't have one but your observation is probably correct.

[Anak]

Originally Posted by GeoffR

A firmware bug is a fault in the code.

A lot of the problems people report are not actually bugs, I agree: some are design choices made by Kobo, some are caused by hardware faults, some are misunderstanding about how something is supposed to work, some by publisher mistakes in the book, etc.

I am fairly sure the problem with the FullScreenReading refresh is caused by a bug.

True, but I don't care if it is actually bug or other issue. It usually ends up with a discussion about semantics. IMO, important is is if it can be replicated and does a function "work" or behave like intended (yes, I know that alone can trigger a discussion; and it always does).
The only exception here is poorly formatted or coded books by the publisher. That is out of Kobos control. The rest isn't. It is Kobos choice to fix or not to fix issues.

Originally Posted by Rand Brittain

It seems to me that complaining about bugs in unsupported features is one of the best ways to get those features up to "supported" status.

I for one really appreciate the fullscreen feature and I'm annoyed that my new H20 arrived just too late for me to avoid installing 3.18 until the problem was fixed.

I agree, if you don't report issues with unsupported or beta functions they will never ever gonna be fixed.

[davidfor]

Originally Posted by Anak

True, but I don't care if it is actually bug or other issue. It usually ends up with a discussion about semantics. IMO, important is is if it can be replicated and does a function "work" or behave like intended (yes, I know that alone can trigger a discussion; and it always does).

Except that what it is has an impact on the likelihood of change. A bug should get fixed. The more severe the bug, the more likely it will be fixed. If it is working as intended, but just not how the reporter wants it to, a change is unlikely.

The only exception here is poorly formatted or coded books by the publisher. That is out of Kobos control. The rest isn't. It is Kobos choice to fix or not to fix issues.

I agree, if you don't report issues with unsupported or beta functions they will never ever gonna be fixed.

Just remember that the fix might be to remove the function.

[Stulle]

Originally Posted by thepossum

Also don't knock the web browser. I have used it to download .epub files directly to my device when I had no usb cable available.

Hi!

I'm also storing my ebooks on dropbox, so I can access and load them directly via wifi to the Kobos if I like.

With 3.18.0 I'm not able to download the files, tapping on a certain file does not show any reaction except a "blinking" (page refresh).
In earlier version (especially 2.x.x) this worked fine, epubs were downloaded and stored on the kobo.
Looks like also current 3.18.0 fw does not follow _any_ external links (e.g. google search results). Tapping on them did'nt work. But manoeuvring throgh a site (e.g. dropbox login, dropbox folder structure) works.
I'm not sure if this just happened with 3.18.0 as I have skipped most earlier 3.x.x. releases

Any idea if this is caused or can be fixed by some configuration file entry?

(Yes, I know the browser is unsupported beta)

Thanks!

Bye,
Michael

[GeoffR]

Originally Posted by Stulle

I'm also storing my ebooks on dropbox, so I can access and load them directly via wifi to the Kobos if I like.

With 3.18.0 I'm not able to download the files, tapping on a certain file does not show any reaction except a "blinking" (page refresh).

I don't know if was a change in the firmware or a change at dropbox, someone with the earlier firmware would need to check.

However I found that I could still download ePub and PDF files from dropbox on my Glo with firmware 3.18.0 by tapping the "Desktop version" button, selecting the file, and then tapping the download button that appears next to the filename at the top of the list.

Edit: I tested on a Mini with firmware 3.4.1 and it is exactly the same (can only download from the desktop version of the site), so I think it must be a dropbox change rather than a firmware change.