Page 1 of 1
PDF help
Posted: Sun Nov 02, 2025 1:17 pm
by Piero
The PDF documentation "formatting" is still horrible
Just an example: # P B _ C o m p i l e r _ P r o c e d u r e (you won't find it…)
Re: PDF help
Posted: Sun Nov 02, 2025 4:06 pm
by Quin
PDF is a garbage format. I hate it.
The reason the text is all split like that is becausew unlike every other sane file format on the planet, PDF just tells you where characters should be rendered! Which is incredibly fun if you're a blind person trying to write a PDF reader, let me tell you.
Anyways, enough OT from me for now.
Re: PDF help
Posted: Sun Nov 09, 2025 12:00 pm
by Piero
Quin wrote: Sun Nov 02, 2025 4:06 pm
PDF is a garbage format. I hate it.
OMG!!! I didn't dare to even try to imagine how it can be for a blind person, if the "reading" is made on "the same results you get when you copy/paste"…
FREEEED!

Re: PDF help
Posted: Tue Nov 11, 2025 9:46 am
by Piero
Quin wrote: Sun Nov 02, 2025 4:06 pm
PDF is a garbage format. I hate it.
Dear Quin, you made me remember that LONG time ago I worked a LOT with PDFs…
It was good and extremely bad at the same time…
Extremely bad:
We had to Print, but the PDFs that customers sent us were often THE WORST NIGHTMARE YOU CAN IMAGINE.
We used a "plugin" called Pitstop (seems it
still exists) and another I cannot remember now, but let's call it "ColorFixer"… and did miracles

Good:
When I went to "other printing places", after showing Pitstop and "ColorFixer" to the poor printing ppl, they looked at me like if I had just discovered fire and they had to throw rose petals for me to step on
I'm out of that since long ago, but I hoped/thought that in 2025 a FREEWARE thing to fix PDFs, or to "digitally print" them decently, existed…
I don't want to search for it; it would make me vividly remember the old nightmare PDFs
FREEEED!
Re: PDF help
Posted: Fri Nov 14, 2025 11:58 am
by Piero
I would be very curious to know exactly what "Fred" used to "compose" the PDF help.
Possibly with "a downloadable example file" for that program, like "
page 293" (294): "UserGuide - Constants"…
Try copy/paste that… and what are those ?? indents?
I'm hoping it to be easily fixable, or exportable to a better free program… (Apple Pages? Calibre? Libre Office? AsciidocFX?) …
ON MAC!

Re: PDF help
Posted: Tue Nov 18, 2025 12:27 pm
by Piero
Grrr… I was 99% sure it was a "font not embedded" problem, but seems it's not!
Wish I also had HomeBrew and shell on Mac when I was working with PDFs…:
Code: Select all
#> pdffonts PureBasic.pdf
name type emb sub uni prob object ID
---------------------------------------------- ----------------- --- --- --- ---- ---------
GNFPVM+SFRM1200 Type 1 yes yes no 7989 0
PBYRSZ+SFRM1728 Type 1 yes yes no 7988 0
MHZLCJ+SFRM2074 Type 1 yes yes no 7987 0
DMKPXF+SFRM1000 Type 1 yes yes no 8024 0
SSMGZN+SFBX2488 Type 1 yes yes no 8021 0
VPXGYX+SFBX1200 Type 1 yes yes no 8022 0
ETYKLU+SFBX1000 Type 1 yes yes no 8023 0
XFEHME+SFBX2074 Type 1 yes yes no 10147 0
ZOQJMF+SFTT1000 Type 1 yes yes no 10159 0
KRHULS+CMSY10 Type 1 yes yes no X 10278 0
SUBKLU+SFIT1000 Type 1 yes yes no 12933 0
VNHPZN+CMSY7 Type 1 yes yes no X 16207 0
LKVDUG+SFBX1440 Type 1 yes yes no 20183 0
RKHEPQ+CMR7 Type 1 yes yes no X 47159 0
ZLBWQW+CMSY8 Type 1 yes yes no X 52920 0
I don't want to install adobe acrobat to double-check… it seems to be a weird Latex/Unicode problem…
At least TRY to EXPORT (or print…) to PDF on a Mac!
PS: just for fun, I exported all pages to jpg and then OCRed them back to a PDF (wasn't freeware; freeware gave worse results…)
I lost "internal links/bookmarks", but on new PDF searching/copying is "better"…
PPS:
Code: Select all
uni "yes" if there is an explicit "ToUnicode" map in the PDF file
(the absence of a ToUnicode map doesn't necessarily mean that
the text can't be converted to Unicode)
prob "X" if this font is likely to be problematic when converting
text to Unicode
ALSO TRY CHANGE FONTS TO STANDARD (WEB) ONES!
PPPS:
Thanks! While criticizing you, I learnted nu sh*t!

Re: PDF help
Posted: Tue Nov 18, 2025 1:29 pm
by Piero
IMPORTANT NOTE
The paid thing I used to OCR images uses a
FREEWARE, OPENSOURCE OCR engine:
https://github.com/tesseract-ocr/tesseract
The freeware one was Apple Vision