I am using PhantomJs 1.8.1 on Centos 6.3 to for automated ui tests. When a test fails, screenshots are saved to the file system.
My problem is that even though the screenshots are saved, they do not contain readable fonts.
So if the website reads like this:
Hello, World!
the screenshot of the site will look like this:
So, instead of the actual letters, it renders and saves little boxes.
The system is centos 6.3.
Freetype and Fontconfig are also installed.
How could i go about fixing this?
Thx!
I had the same problem.
Installing the urw-fonts package solved it for me:
yum install urw-fonts
I had a similar problem with Japanese fonts. (PhantomJS 1.9.1, Redhat on Amazon EC2)
English characters showed up fine, but Japanese characters were rendered as boxes.
How I fixed it:
1) Installed the (Japanese) IPA fonts (Mincho and Gothic) using yum install.
(Use yum list to check the exact package names.)
2) The IPA .ttf files were installed to:
/usr/share/fonts/IPA-Gothic/
/usr/share/fonts/IPA-Mincho/
3) Move the two downloaded .ttf files to this directory: (Create it)
/usr/share/fonts/ipa/
4) Make a backup of /etc/fonts/fonts.conf
5) Edit the original /etc/fonts/fonts.conf and fill it with this:
<?xml version="1.0"?>
<!DOCTYPE fontconfig SYSTEM "fonts.dtd">
<fontconfig>
<dir>/usr/share/fonts/ipa</dir>
<cachedir>/var/cache/fontconfig</cachedir>
<cachedir>~/.fontconfig</cachedir>
<alias>
<family>serif</family>
<prefer>
<family>IPAP Mincho</family>
</prefer>
</alias>
<alias>
<family>sans serif</family>
<prefer>
<family>IPAP Gothic</family>
</prefer>
</alias>
<alias>
<family>monospace</family>
<prefer>
<family>IPA Gothic</family>
</prefer>
</alias>
</fontconfig>
6) Refresh your font cache with fc-cache -vf
7) Enjoy your new working fonts.
Gotchas:
If you're getting no characters (blank space), your font cache is probably out of date.
Try fc-cache -vf to regenerate it.
There's a fix for Japanese/Chinese/Korean characters in the 1.9.1 release. Not sure if it makes a difference, but probably worth upgrading from 1.9.0.
For Chinese font, I had it solved by the following steps:
sudo apt-get install language-pack-zh-hans
sudo apt-get install ttf-arphic-uming
sudo apt-get install ttf-dejavu ttf-wqy-microhei
sudo fc-cache -f -v
OS is Ubuntu 12.04 LTS
I have the same problem on amazon ec2
I fix it by this:
yum install cjkuni-ukai-fonts
Also you can try instaling dependencies - FontConfig & FreeType
yum install fontconfig
yum install freetype*
You can run the script with command line parameters:
phanthomjs --output-encoding=cp866 [params] [filename]
I faced same issue with Arabic fonts. This is what i did.
yum groupinstall 'Arabic Support'.
Installing Arabic Support solved it for me.
Try this for Chinese.
yum install bitmap-fonts bitmap-fonts-cjk
leave the /etc/fonts/fonts.conf file alone. the example above is making only Japanese work
yum install -y ipa-gothic-fonts.noarch ipa-mincho-fonts.noarch cjkuni-ukai-fonts.noarch bitmap-fonts bitmap-fonts-cjk urw-fonts fontconfig freetype*
yum groupinstall -y 'Korean Support' 'Chinese Support' 'Japanese Support' 'Kannada Support' 'Hindi Support' 'Arabic Support'
fc-cache -vf
Add a style section to your HTML section Something like this:
<style type="text/css">
#font-face
{
font-family:MyFont;
src: url('MyFont.ttf') format('truetype');
}
#barcodefont
{
font-family:MyFont;
font-size: 42px;
color:black;
}
</style>
Then to use the font in the main HTML code do this:
<div id="MyFont">Your text using MyFont</div>
For this to work, your HTML file and your MyFont.ttf file both have to be in the PhantomJS directory that you are doing the conversion.
Related
Problem
I need to convert a multipage pdf to jpg-files but ImageMagick keeps throwing errors that are hard to interpret.
Installing ImageMagick
At first I installed it using apt-get, but as I could read that several people had problems doing that, i ended up installing it from source.
My linux distribution (A Docker image):
>lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description: Debian GNU/Linux 10 (buster)
Release: 10
Codename: buster
Installing ImageMagick from source:
# Installing build tools and ghostscript
apt update
apt-get install -y build-essential make ghostscript
# Downloading imagemagick
wget https://www.imagemagick.org/download/ImageMagick.tar.gz
# Installing and cleaning up
tar xvzf ImageMagick.tar.gz && cd ImageMagick-7* && ./configure && make && make install && ldconfig /usr/local/lib && cd .. && rm -r ImageMagick-7*
# Checking ImageMagick version
>magick -version
Version: ImageMagick 7.0.10-60 Q16 x86_64 2021-01-25 https://imagemagick.org
Copyright: (C) 1999-2021 ImageMagick Studio LLC
License: https://imagemagick.org/script/license.php
Features: Cipher DPC HDRI OpenMP(4.5)
Delegates (built-in): jpeg x xml zlib
Converting files
# Image to image
>convert test.jpg test.png
# PDF to image
>convert test.pdf test.jpg
convert: no decode delegate for this image format `' # error/constitute.c/ReadImage/572.
convert: no images defined `test.jpg' # error/convert.c/ConvertImageCommand/3304.
Is Ghostscript the problem?
The Ghostscript installation is a common problem for many, but Ghostscript seems to work fine and produce a jpg-file
# PDF to image with Ghostscript
gs -sDEVICE=pngalpha -sOutputFile=test.jpg test.pdf
Do I have to install more Delegates?
The error suggests that there is something off with my limited delegates, so I thought to install all dependencies up front.
# Listing dependencies
>apt update && apt build-dep imagemagick
Reading package lists... Done
E: You must put some 'source' URIs in your sources.list
This is where I got stuck.
Solution
It turned you that it was indeed the delegates that were missing. I haven't seen this well described in the documentation or anywhere else.
NOTE: Delegates should be installed before you install ImageMagick
Here is how I fixed it:
# Add source URI or uncomment source URI
## Adding URI
echo "deb-src http://deb.debian.org/debian buster main" >> /etc/apt/sources.list
apt update
## Uncommenting URI
sudo sed -Ei 's/^# deb-src /deb-src /' /etc/apt/sources.list
sudo apt update
# Installing dependencies
apt-get build-dep imagemagick
Now I can convert pdf to jpg!
My solution was to reinstall imagemagick, but following the steps in the answer of andrew.46 at https://askubuntu.com/questions/745660/imagemagick-png-delegate-install-problems/746195#746195
Followed this link to try and generate pdf from Sphinx:
https://www.quora.com/How-to-create-a-PDF-out-of-Sphinx-documentation-tool
$ sphinx-build -b pdf source build/pdf
Error: Cannot find source directory `/Users/seb/mydocs/source'.
$ make all-pdf
make: *** No rule to make target `all-pdf'. Stop.
$ make pdf
make: *** No rule to make target `pdf'. Stop.
Since tried in OSX:
$ conda install -c dfroger rst2pdf=0.93
Fetching package metadata .........
Solving package specifications: .
Error: Package missing in current osx-64 channels:
- rst2pdf 0.93*
You can search for packages on anaconda.org with
anaconda search -t conda rst2pdf
EDIT:
After pip install rst2pdf
install rst2pdf
register rst2pdf in your conf.py Sphinx config
extensions = ['sphinx.ext.autodoc','rst2pdf.pdfbuilder']
But adding 'rst2pdf.pdfbuilder' causes
Extension error:
Config value 'math_number_all' already present
make: *** [html] Error 1
$ sphinx-build -bpdf sourcedir outdir
But what do I specify as sourcedir and outdir? Example please.
EDIT:
Now after make html
and then:
$ rst2pdf index.rst output.pdf
index.rst:14: (ERROR/3) Unknown directive type "toctree".
.. toctree::
:maxdepth: 2
introduction
tutorial
multiple_jobs
deployment
project
index.rst:26: (ERROR/3) Unknown interpreted text role "ref".
index.rst:27: (ERROR/3) Unknown interpreted text role "ref".
index.rst:28: (ERROR/3) Unknown interpreted text role "ref".
Also:
$rst2pdf.py index.rst -o mydocument.pdf
Does produce a mydocument.pdf but completely different from html and toc to all the pages are not even there?
Image of pdf verse HTML same page
This is from the official Sphinx documentation. If you have pdfTex tool installed in your machine, all you need is:
$ make latexpdf
Then, the generated pdf file(s) will be under _build/latex/<PROJECT-NAME>.pdf
So, the complete process from scratch would be as follows:
$ pip install -U sphinx # install the package
$ sphinx-quickstart # create a new project (answer the questions)
$ make latexpdf # compile and generate pdf file
Note that you may also "optionally" install whatever extensions needed by editing the file config.py
NOTE: This answer assumes LaTeX engine is installed on your machine.
I have succeeded in generating a PDF file for the DevStack document by following the configuration changes in this link:
Here are the steps:
Edit your conf.py (edit or append values)
extensions = ['rst2pdf.pdfbuilder']
pdf_documents = [('index', u'rst2pdf', u'Sample rst2pdf doc', u'Your Name'),]
Install the "rst2pdf" if necessary
pip install rst2pdf
Build the PDF file like this:
sphinx-build -b pdf doc/source doc/build
Succeeded in Pdf Generation via Latex (for windows 10)... No need to change existing conf.py file in Sphinx... the best solution is install MiKTeX....install Perl (ActiveState).... after when running sphinx type 'make HTML' type 'make latex' and then make latexPdf... this solved my issue.
You could avoid rst2pdf and use make pdflatex to build pdf output via a latex file.
cf more info:
https://docs.typo3.org/typo3cms/extensions/sphinx/AdvancedUsersManual/RenderingPdf/
Make sure to have a look at http://rst2pdf.ralsina.me/handbook.html#sphinx.
Following applies to Ubuntu 16.
But it is probably less painful to install a full LaTex suite than to try to get this tool running; it is very sensitive to errors, and is difficult to use.
I took a look there (https://tex.stackexchange.com/a/95373) and it looks daunting...
this is more encouraging: https://milq.github.io/install-latex-ubuntu-debian/
And I did it:
sudo apt-get install texlive-full
make clean latexpdf
is worthwhile your patience (and disk space) to install. I got rid of rst2pdf.
Found this related post which discovers that rst2pdf breaks math rendering in Sphinx 1.4.1 because it imports a dummy math module.
In rst2pdf.pdfbuilder:setup(), it calls mathbase.setup() internally to install dummy math module. It causes conflicts and raise errors.
I had this same error when running sphinx-build -b pdf
Extension error:
Config value 'math_number_all' already present
Now looking at latex option, and exporting to pdf as #tfv posted. pdflatex is a larger universe of TeX distribution.
As an alternative to latexpdf: rinohtype
Once rinohtype is installed, sphinx-build -b rinoh . _build
Create a file documentation.md, and open it in Typora software. Navigate to the Sphynx documentation that you need, mark with mouse and copy inside Typora. Then inside typora export as PDF. Done.
I have just upgraded my Macbook Pro OS to El Capitan (v10.11.4).
My attempt to export a Markdown file (created using Sublime Text 2, v2.0.2, build 2221) to pdf using pandoc is now failing, and I receive the following error:
pandoc: xelatex not found. xelatex is needed for pdf output
My output command is as follows:
pandoc doc1.md -o doc1.pdf --toc -V geometry:margin=1in --variable fontsize=10pt --variable fontfamily=utopia --variable linkcolor=blue --latex-engine=xelatex -f markdown-implicit_figures -s
Above command worked like a charm prior to installing El Capitan.
FYI - in searching for questions here I have not found one that gives a suitable answer.
For my case, add one line into ~/.bashrc solved the error:
export PATH=/Library/TeX/texbin:$PATH
Of course, the environment variable should be activated in the current term:
$ . ~/.bashrc
then run: $ make
the error disappears.
El Capitan's security features disable and remove the old symlink /usr/texbin. If you have MacTeX 2015, they should've been installed in /Library/TeX/texbin as well. You'll have to update the PATH your using to launch pandoc to include that folder. If you have a pre-2015 distribution of MacTeX, there are instructions here.
Linux Ubuntu instructions:
Tested on Ubuntu 18.04:
If you see this error on Linux Ubuntu:
pandoc: xelatex not found. xelatex is needed for pdf output
Then you need to install the texlive-xetex package like this:
sudo apt update
sudo apt install texlive-xetex
That solves it! Source where I learned this: TEX: XeLatex under Ubuntu.
In my particular case, I was trying to run this make_book.sh script to generate book.pdf, so I needed to do all of the following:
sudo apt update
sudo apt install pandoc
pip3 install MarkdownPP
sudo apt install texlive-xetex
cd path/to/repo
cd systemd-by-example
./make_book.sh
# You'll now have "book.pdf" inside directory "systemd-by-example"!
References:
https://github.com/jreese/markdown-pp - instructions to install MarkdownPP
https://tex.stackexchange.com/a/179811/168682 - instructions to install texlive-xetex
I have a problem with this module.
In my local server i open my php.ini and i have the module imagick in the list.
Now i changed the site to a webserver, but in php.ini, the module is not showed.
I talked with the company that have the web server, and the answer is: "the module is installed and show me this:"
root#dime38 [~]# convert
Version: ImageMagick 6.2.8 08/25/10 Q16 file:/usr/share/ImageMagick-6.2.8/doc/index.html
but when i use this code:
<?php
$image = new Imagick();
$image->newImage(100, 100, new ImagickPixel('red'));
$image->setImageFormat('png');
header('Content-type: image/png');
echo $image;
?>
i simple receive this:
Fatal error: Class 'Imagick' not found in /home/empreg0l/public_html/modulo.php on line 3
But the same code works in my local host.
What is the problem? (Probably, the extension is commented in the php.ini? or exists any problem in the code?)
thanks
There is a difference between the ImageMagick binary (which can be called through the convert command) and the IMagick PHP extension. Even if the binary is installed, it doesn't mean the PHP extension is.
Your provider would have to explicitly activate that in their server's PHP.
If they won't do that, you'll have to recreate the IMagick commands as command line options and call it through exec().
To install imagick for php:
apt-get install php5-imagick
Use get_loaded_extensions to confirm you have the imagick PHP extension installed.
var_dump(get_loaded_extensions());
apt-get install pkg-config libmagickwand-dev -y
cd /tmp
wget https://pecl.php.net/get/imagick-3.4.0.tgz
tar xvzf imagick-3.4.0.tgz
cd imagick-3.4.0
phpize
./configure
make install
rm -rf /tmp/imagick-3.4.0*
echo extension=imagick.so >> /etc/php/7.0/cli/php.ini
echo extension=imagick.so >> /etc/php/7.0/fpm/php.ini
From Install Imagick 3.4.0 on PHP 7.0
I have tried in all ways to get wkhtmltopdf installed on our web server but unfortunately it is not getting installed. I cannot access user/bin folder as stated in a tutorial on installation.
On the server in public_html folder there is a sub folder _vti_bin, I copied the file wkhtmltopdf-i386 from wkhtmltopdf-0.9.1-static-i386, but I am not able to execute it.
How to install wkhtmltopdf on (shared hosting) web server and get it working?
I've managed to successfully install wkhtmltopdf-amd64 on my shared hosting account without root access.
Here's what i did:
Downloaded the relevant static binary v0.10.0 from here: http://code.google.com/p/wkhtmltopdf/downloads/list
EDIT: The above has moved to here
via ssh on my shared host typed the following:
$ wget {relavant url to binary from link above}
$ tar -xvf {filename of above wget'd file}
you'll then have the binary on your host and will be able to run it regardless of if its in the /usr/bin/ folder or not. (or at least i was able to)
To test:
$ ./wkhtmltopdf-amd64 http://www.example.com example.pdf
Note remember that if you're in the folder in which the executable is, you should probably preface it with ./ just to be sure.
Worked for me anyway
If you have sudo access...
Ubuntu 14.04 / 15.04 / 18.04:
sudo apt-get install wkhtmltopdf
# or
sudo apt install wkhtmltopdf
Others
Look at the other answers.
If its ubuntu then go ahead with this, already tested.:--
first, installing dependencies
sudo aptitude install openssl build-essential xorg libssl-dev
for 64bits OS
wget http://wkhtmltopdf.googlecode.com/files/wkhtmltopdf-0.9.9-static-amd64.tar.bz2
tar xvjf wkhtmltopdf-0.9.9-static-amd64.tar.bz2
mv wkhtmltopdf-amd64 /usr/local/bin/wkhtmltopdf
chmod +x /usr/local/bin/wkhtmltopdf
for 32bits OS
wget http://wkhtmltopdf.googlecode.com/files/wkhtmltopdf-0.9.9-static-i386.tar.bz2
tar xvjf wkhtmltopdf-0.9.9-static-i386.tar.bz2
mv wkhtmltopdf-i386 /usr/local/bin/wkhtmltopdf
chmod +x /usr/local/bin/wkhtmltopdf
Debian 8 Jessie
This works
sudo apt-get install wkhtmltopdf
Chances are that without full access to this server (due to being a hosted account) you are going to have problems. I would go so far as to say that I think it is a fruitless endeavor--they have to lock servers down in hosted environments for good reason.
Call your hosting company and make the request to them to install it, but don't expect a good response--they typically won't install very custom items for single users unless there is a really good reason (bug fixes for example).
Lastly, depending on how familiar you are with server administration and what you are paying for server hosting now consider something like http://www.slicehost.com. $20 a month will get you a low grade web server (256 ram) and you can install anything you want. However, if you are running multiple sites or have heavy load the cost will go up as you need larger servers.
GL!
Latest update for CentOS:
sudo yum install -y libpng libjpeg openssl icu libX11 libXext libXrender xorg-x11-fonts-Type1 xorg-x11-fonts-75dpi
wget https://github.com/wkhtmltopdf/wkhtmltopdf/releases/download/0.12.4/wkhtmltox-0.12.4_linux-generic-amd64.tar.xz
tar -xvf wkhtmltox-0.12.4_linux-generic-amd64.tar
sudo mv wkhtmltox/bin/* /usr/local/bin/
check installation success: wkhtmltopdf -V
rm -rf wkhtmltox
rm -f wkhtmltox-0.12.4_linux-generic-amd64.tar
Place the wkhtmltopdf executable on the server and chmod it +x.
Create an executable shell script wrap.sh containing:
#!/bin/sh
export HOME="$PWD"
export LD_LIBRARY_PATH="$PWD/lib/"
exec $# 2>/dev/null
#exec $# 2>&1 # debug mode
Download needed shared objects for that architecture and place them an a folder named "lib":
lib/libfontconfig.so.1
lib/libfontconfig.so.1.3.0
lib/libfreetype.so.6
lib/libfreetype.so.6.3.18
lib/libX11.so.6 lib/libX11.so.6.2.0
lib/libXau.so.6 lib/libXau.so.6.0.0
lib/libxcb.so.1 lib/libxcb.so.1.0.0
lib/libxcb-xlib.so.0
lib/libxcb-xlib.so.0.0.0
lib/libXdmcp.so.6
lib/libXdmcp.so.6.0.0
lib/libXext.so.6 lib/libXext.so.6.4.0
(some of them are symlinks)
… and you're ready to go:
./wrap.sh ./wkhtmltopdf-amd64 --page-size A4 --disable-internal-links --disable-external-links "http://www.example.site/" out.pdf
If you experience font problems like squares for all the characters, define TrueType fonts explicitly:
#font-face {
font-family:Trebuchet MS;
font-style:normal;
font-weight:normal;
src:url("http://www.yourserver.tld/fonts/Trebuchet_MS.ttf");
format(TrueType);
}
List of stable versions wkhtmltopdf: http://wkhtmltopdf.org/downloads.html
Installing wkhtmltopdf on Debian 8.2 (jessie) x64:
sudo apt-get install xfonts-75dpi
sudo apt-get install xfonts-base
sudo wget http://download.gna.org/wkhtmltopdf/0.12/0.12.2.1/wkhtmltox-0.12.2.1_linux-jessie-amd64.deb
sudo dpkg -i wkhtmltox-0.12.2.1_linux-jessie-amd64.deb
Shared hosting no ssh or shell access?
Here is how i did it;
Visit https://wkhtmltopdf.org/downloads.html and download the appropriate stable release for Linux. For my case I chose 32-bit
which is wkhtmltox-0.12.4_linux-generic-i386.tar.xz
Unzip to a folder on your local drive.
Upload the folder to public_html (or whichever location fits your need) using an FTP program just like any other file(s)
Change the binary paths in snappy.php file to point the appropriate files in the folder you just uploaded.
Bingo! there you have it. You should be able to generate PDF files.
A few things have changed since the top answers were added. They used to work out for me, but not quite anymore, so I have been hacking around for a bit and came up with the following solution for Ubuntu 16.04. For Ubuntu 14.04, see the comment at the bottom of the answer. Apologies if this doesn't work for shared hosting, but it seems like this is the goto answer for wkhtmltopdf installation instructions in general.
# Install dependencies
apt-get install libfontconfig \
zlib1g \
libfreetype6 \
libxrender1 \
libxext6 \
libx11-6
# TEMPORARY FIX! SEE: https://github.com/wkhtmltopdf/wkhtmltopdf/issues/3001
apt-get install libssl1.0.0=1.0.2g-1ubuntu4.8
apt-get install libssl-dev=1.0.2g-1ubuntu4.8
# Download, extract and move binary in place
curl -L -o wkhtmltopdf.tar.xz https://github.com/wkhtmltopdf/wkhtmltopdf/releases/download/0.12.4/wkhtmltox-0.12.4_linux-generic-amd64.tar.xz
tar -xf wkhtmltopdf.tar.xz
mv wkhtmltox/bin/wkhtmltopdf /usr/local/bin/wkhtmltopdf
chmod +x /usr/local/bin/wkhtmltopdf
Test it out:
wkhtmltopdf http://www.google.com google.pdf
You should now have a file named google.pdf in the current working directory.
This approach downloads the binary from the website, meaning that you can use the latest version instead of relying on package managers to be updated.
Note that as of today, my solution includes a temporary fix to this bug. I realize that the solution is really not great, but hopefully it can be removed soon. Be sure to check the status of the linked GitHub issue to see if the fix is still necessary when you read this answer!
For Ubuntu 14.04, you will need to downgrade to a different version of libssl. You can find the versions here. Anyways, be sure to consider the implications of downgrading libssl before doing so on any production server.
I hope this helps someone!
After trying, below command work for me
cd ~
yum install -y xorg-x11-fonts-75dpi xorg-x11-fonts-Type1 openssl git-core fontconfig
wget https://downloads.wkhtmltopdf.org/0.12/0.12.4/wkhtmltox-0.12.4_linux-generic-amd64.tar.xz
tar xvf wkhtmltox-0.12.4_linux-generic-amd64.tar.xz
mv wkhtmltox/bin/wkhtmlto* /usr/bin
Version 12.5 of wkhtmltopdf only lists DEB files on their download page now. Being a mac user and not knowing much linux or what DEB files were I couldn't use the solutions posted.
This page helped me get past the knew twist of downloading a DEB file: http://www.g-loaded.eu/2008/01/28/how-to-extract-rpm-or-deb-packages/
Basically what I did was:
Downloaded from https://wkhtmltopdf.org/downloads.html
Unzipped the DEB file.
Unzipped data.tar.xz
Uploaded the binary in the unzipped 'usr' folder from step 3 (usr/local/bin/wkhtmltopdf)
Then I found out that the 'exec' function was disabled on my host. So make sure you can specifically run 'exec' if you're using PHP to run this. "Can I run the wkhtmltopdf binary" isn't specific enough. My fault.