PDF is Driving me Crazy !!

[kovidgoyal]

Although I grew up with the GUI the command line just seems more natural. TO me, its like talking to the computer as opposed to miming (i.e. pointing at things) which is what you do with a GUI.

[mogui]

I opened a Microsoft Word "doc" file in OpenOffice. I resized the pages throughout the document (there were 3 sections) to 3.57" x 4.82" (90.6mm x 122.4mm). I exported the file as a PDF.

When opened with Adobe Acrobat and viewed "Actual Size", the PDF document looked appropriately sized for the Sony Reader.

When transferred to the Reader, the pages looked just fine. It appeared they could be displayed in two sizes, though the sizes were not too dissimilar. They were also quite readable in landscape mode.

It seems that, if desired, PDFs can be created easily by this method.

Originally Posted by nerys

I want to convert some HTML files to PDF

Next I used OpenOffice to load an HTML file. I changed the page format to the above measurements and exported the file as a PDF document. This file contained images. The result was usable and readable on the Sony Reader. The pages fit the Reader screen as expected.

[Steven Lyle Jordan]

MS Word on Windows will do the same (I'm assuming so will Word on Mac). With other word processing apps that do not directly import or convert HTML, you may need to convert the HTML to a TXT or RTF file first, then import into the WP app. Then you can resize as needed (See Mogui's instructions).

[nerys]

I tried open office and it does work (set page size then export to PDF) but I loose all the colors and images :-(

Plus it made it a lot larger (27 pages to 54 pages) so I just used IMAGES. I captured a image of the whole thing chopped it up just as I wanted it and made those into a PDF I reduced it all to 4color grey first so it was already pretty optimized and still very tiny. Worked a treat. was just hoping for a bit more automation.

[Steven Lyle Jordan]

I know the feeling... every time I get ready to do a book conversion to seven different files for the SJB, I always have to shake off a shiver of angst first!

[SoCal Bob]

Originally Posted by nerys

I tried open office and it does work (set page size then export to PDF) but I loose all the colors and images :-(

Plus it made it a lot larger (27 pages to 54 pages) so I just used IMAGES. I captured a image of the whole thing chopped it up just as I wanted it and made those into a PDF I reduced it all to 4color grey first so it was already pretty optimized and still very tiny. Worked a treat. was just hoping for a bit more automation.

You need to define a page size as "Sony Reader" See this thread:
https://www.mobileread.com/forums/sho...3851#post53851
This is how I did it in Windows then you can open a new document in this size and paste any HTML, graphs, pictures, etc, you will then be able to print the correct size in PDF for the reader.

[SoCal Bob]

BTW here is a paste up of a brief article and a graph from the Economist yesterday. See if this works for you and on your reader

[alex_d]

OMG! PAGE SIZE DOESN'T MATTER AT ALL!

Why do people keep worrying so much about it, writing conversion guides specifying it to 4 digits of precision?

All that matters is the page shape (w:h = 0.75, and 8.5":11" is good enough) and font size.

If you double the page size and double the font size, the document will look exactly the same. You people are crazy.

P.S. kovigdoyal... command-line is like talking to the computer? Pass me your joint. Even if i play along, (and the analogy is rather apt), using the command-line is more like trying to speak a backward language you barely know and that no one is teaching you. Some phrases you've memorized in entirety, a number of sentences you can piece together, but for the majority of syntax you have to stop and take out a stupid dictionary. But that only works if you already know what you want to say.

No matter how good you claim you are at unix or dos, you're still extremely fundamentally limited (although you probably don't realize it, because you do know how to do what you know how to do). With a GUI, though, I can open up any program I have never used in my life and get things done. More importantly, I actually do this every week because it's easy and convenient. In programs i've used for a long time it's easier to venture out and explore functionality i've previously overlooked. With the command-line, the more you use a program the less chance you're going to learn how to use it any differently (to say nothing of switching to a different program that does the same thing better). That's why command-line tools are still the same as they were 20 years ago, while GUI tools continue to evolve at lightning speed.

I do like the keyboard, however. But, it's easier to use the keyboard with a GUI than with a command-line. Take the simple task of navigating to a file using Windows Explorer vs BASH. I can use the keyboard to type folder and file names in both, but in windows explorer they highlight automatically (in bash I have to keep pressing TAB like a monkey), and I can right away see all the filenames starting from those characters (in bash, double TAB). I can also see all the filenames which DON"T start from those characters (in bash, forget it. you have to start a new command and fire 'ls'). Not to mention I automatically see file size, modification date, and other attributes (in bash, more 'ls' commands). If I want to move two files with distinct names, I can do it using CTRL and SPACE (in bash, you have to retype each filename from scratch). And should I need it, I can always just pull out the mouse. Oh, and what if I'd like to see thumbnails? ALT+V+H. What if I forget this combination? There's underlined letters and it's 100x superior to having to do 'man.' BASH, Linux, Unix, POSix is pathetic in comparison. But it's too bad most users don't know the power of the keyboard in Windows.

[nerys]

Would be nice if you knew what you were talking about.

YES it will scale and be nicely unreadable on the smaller screen when its scaled down. Yeah thats great :-) really not flaming just being funny

its an HTML file. I want to make it the PROPER size so I can then "edit" it to "fit" the information as effeciently as possible. Put page breaks where its most effecient. Since HTML "flows" based on the size of the canvas I need the canvas to be the FINAL size in order to properly edit the document.

BUT I want to retain all the shading and colors etc..

I can drag and drop an HTML into Adobe Acrobat but it will not let me DEFINE the page it generates when I do this. its always 8.5x11

I can generate the A5 page in say open office and then import into acrobat but I loose all the images and color and shading etc..

I ended up loading it live in a web browser condensing it as effeciently as I could and then capturing an image of the entire thing and "chopping" it up into pages. reduced to 4color grey it ended up very small so I was happy. Was just hoping to avoid all that work.

[nerys]

GUI verse Command Prompt.

My reason for hating the command prompt is simple. I spend 8+ hours a day staring at a stinking screen. I can stare at pretty pictures and graphics or a stinking text prompt. Guess which I prefer.

I want it to be PRETTY and NICE prompts suck for that Gui's rock for that. :-)

[alex_d]

Nerys, I can see one reason for changing the page size rather than font size: if there are already many fonts in different sizes and there's no command to make them all larger by the same amount, then just making the page smaller is easier.

However, the stuff about page breaks, readability, formatting, some stupid notion about the "proper" size that a pdf page should be (112.4mm to the angstrom, dammit!) is completely wrong. Everything works identically if you change the font size rather than page size. PDFs don't have problems being "scaled"! They are vector-based. They scale perfectly.

As for converting html to pictures... hmm, sounds like a cool thing for RasterFarian to do. Anyone know a command-line program for rendering html into images?

[NatCh]

Originally Posted by alex_d

With a GUI, though, I can open up any program I have never used in my life and get things done.

-- within the limits of what the GUI has been programmed to do, yes. If the GUI hasn't been programmed to allow you do do something then you're just not going to be able to do it within the GUI.

GUI's are friendlier and prettier (well, usually) than command lines, but they're only more powerful to the limits of the skill of the programmer who created them.

[HarryT]

Originally Posted by alex_d

Nerys, I can see one reason for changing the page size rather than font size: if there are already many fonts in different sizes and there's no command to make them all larger by the same amount, then just making the page smaller is easier.

If you're using "Word" there is such a command, of course: "Ctrl+Shift+>" will make everything in the selected region 1pt larger, "Ctrl+Shift+<" 1 pt smaller. The increment multiplies if you use the command multiple times.

[kovidgoyal]

That's the silliest justification of GUI I've ever read. I seriously doubt you need my joint, you seem to have a more than adequate supply of your own.

By command line language I mean a scripting language you can use on the command line, for e.g. I use ipython as my system shell which lets me use the full expressive power of python at the command line. Bash script is pretty powerful as well though it's a little awkward for more complex tasks. So the command line language is Turing complete and if you take the time to learn it as well as you learn a human language (and it's a lot easier to do) you will realize just what I'm talking about. Know any Turing complete GUIs?

As for using new applications...the only reason you find new GUI applications easier is because you haven't taken the time to really use a commandline and you seem to work in windows...

I'm not going to get into a "proof by example" fight with you. If you want to argue about the relative merits of two user interface paradigms, examples aren't going to cut it.

Originally Posted by alex_d

P.S. kovigdoyal... command-line is like talking to the computer? Pass me your joint. Even if i play along, (and the analogy is rather apt), using the command-line is more like trying to speak a backward language you barely know and that no one is teaching you. Some phrases you've memorized in entirety, a number of sentences you can piece together, but for the majority of syntax you have to stop and take out a stupid dictionary. But that only works if you already know what you want to say.

No matter how good you claim you are at unix or dos, you're still extremely fundamentally limited (although you probably don't realize it, because you do know how to do what you know how to do). With a GUI, though, I can open up any program I have never used in my life and get things done. More importantly, I actually do this every week because it's easy and convenient. In programs i've used for a long time it's easier to venture out and explore functionality i've previously overlooked. With the command-line, the more you use a program the less chance you're going to learn how to use it any differently (to say nothing of switching to a different program that does the same thing better). That's why command-line tools are still the same as they were 20 years ago, while GUI tools continue to evolve at lightning speed.

I do like the keyboard, however. But, it's easier to use the keyboard with a GUI than with a command-line. Take the simple task of navigating to a file using Windows Explorer vs BASH. I can use the keyboard to type folder and file names in both, but in windows explorer they highlight automatically (in bash I have to keep pressing TAB like a monkey), and I can right away see all the filenames starting from those characters (in bash, double TAB). I can also see all the filenames which DON"T start from those characters (in bash, forget it. you have to start a new command and fire 'ls'). Not to mention I automatically see file size, modification date, and other attributes (in bash, more 'ls' commands). If I want to move two files with distinct names, I can do it using CTRL and SPACE (in bash, you have to retype each filename from scratch). And should I need it, I can always just pull out the mouse. Oh, and what if I'd like to see thumbnails? ALT+V+H. What if I forget this combination? There's underlined letters and it's 100x superior to having to do 'man.' BASH, Linux, Unix, POSix is pathetic in comparison. But it's too bad most users don't know the power of the keyboard in Windows.

[kovidgoyal]

Originally Posted by nerys

GUI verse Command Prompt.

My reason for hating the command prompt is simple. I spend 8+ hours a day staring at a stinking screen. I can stare at pretty pictures and graphics or a stinking text prompt. Guess which I prefer.

I want it to be PRETTY and NICE prompts suck for that Gui's rock for that. :-)

Let me introduce you to the concept of using background images in terminals ;-)