Render PDF copies of tldr pages (#1964)

2025-03-28 21:16:20 +01:00 · 2018-03-12 00:23:48 +05:30 · 2018-03-12 00:23:48 +05:30 · 490da5763d
commit 490da5763d
parent 930cfe7f89
6 changed files with 283 additions and 0 deletions
--- a/scripts/pdf/PT_Serif-Web-Regular.ttf
+++ b/scripts/pdf/PT_Serif-Web-Regular.ttf
--- a/scripts/pdf/README.md
+++ b/scripts/pdf/README.md
@ -0,0 +1,48 @@
+# pdf
+
+This directory contains the script and related resources to generate a PDF copy of tldr pages.
+
+## Preview
+
+![aa](https://user-images.githubusercontent.com/29029116/35637791-4e42af80-06db-11e8-8b8e-42ce6c905ff4.jpg)
+![bb](https://user-images.githubusercontent.com/29029116/35637798-51e3784a-06db-11e8-9576-6e57ef5c5c20.jpg)
+![cc](https://user-images.githubusercontent.com/29029116/35637801-54449fce-06db-11e8-93f7-d90cdc34044b.jpg)
+
+## Highlights
+
+- No LaTeX dependencies for generating the PDF.
+
+- 3 available color-schemes- *Basic*, *Solarized Light* and *Solarized Dark*. More can be added easily through CSS.
+
+
+## Requirements
+
+The PDF is generated by first converting the markdown files to HTML, and then rendering those HTML files as PDF. It depends on the following libraries-
+
+#### Python-Markdown
+
+    pip3 install markdown
+    
+#### WeasyPrint
+
+    pip3 install WeasyPrint
+    
+Make sure OS specific dependencies for WeasyPrint are installed by following the instructions [here](http://weasyprint.readthedocs.io/en/latest/install.html).
+
+## Usage
+
+Generating the PDF is as simple as running
+
+    python3 render.py <path-to-pages-directory> -c <color-scheme>
+    
+Complete information about the arguments can be viewed by running 
+
+    python3 render.py --help
+
+The color-schemes that can be specified are
+
+* `solarized-light`
+* `solarized-dark`
+
+    
+
--- a/scripts/pdf/basic.css
+++ b/scripts/pdf/basic.css
@ -0,0 +1,34 @@
+@font-face {
+    font-family: 'PT_Serif-Web-Regular';
+    src: url('PT_Serif-Web-Regular.ttf')  format('truetype');
+}
+
+p {
+    margin-left: 2.5em;
+}
+
+code {
+    color: darkslategrey;
+}
+
+h1, h4, h2, ul {
+    font-family: "PT_Serif-Web-Regular";
+}
+
+h1.titlemain {
+    text-align: center;
+    margin-top: 6em;
+    font-size: 350%;
+}
+
+h4.titlesub {
+    text-align: center;
+    font-size: 120%;
+    color: darkslategrey;
+}
+
+h2.titledir {
+    text-align: center;
+    margin-top: 8.2em;
+    font-size: 300%;
+}
--- a/scripts/pdf/render.py
+++ b/scripts/pdf/render.py
@ -0,0 +1,122 @@
+#!/usr/bin/env python3
+
+#A Python script to generate a single PDF document with all the tldr pages. It works by generating 
+#intermediate HTML files from existing md files using Python-markdown, applying desired formating 
+#through CSS, and finally rendering them as PDF. There is no LaTeX dependency for generating the PDF.
+
+import os
+import sys
+import glob
+import re
+import markdown
+import argparse
+
+from weasyprint import HTML
+
+def main(loc, colorscheme):
+
+    oslist = []
+    allmd = []
+    group = []
+    ap = []
+
+    #Checking correctness of path
+    if not os.path.isdir(loc):
+        print("Invalid directory. Please try again!", file = sys.stderr)
+        sys.exit(1)
+
+    #Writing names of all directories inside 'pages' to a list
+    for os_dir in os.listdir(loc):
+        oslist.append(os_dir)
+
+    oslist.sort()
+
+    #Required strings to create intermediate HTML files
+    header = "<html><head><link rel=stylesheet type=text/css href=" + colorscheme + ".css></head><body>\n"
+    footer = "</body></html>"
+    title_content = "<h1 class=titlemain>tldr pages</h1><h4 class=titlesub>Simplified and community driven man pages</h4></body></html>"
+
+    #Creating title page
+    with open("title.html", 'w') as f:
+        f.write(header + title_content)
+
+    group.append(HTML('title.html').render())
+
+    for operating_sys in oslist:
+
+        i = 1
+
+        #Required string to create directory title pages
+        dir_title = "<h2 class=titledir>" + operating_sys.capitalize() + "</h2></body></html>"
+
+        #Creating directory title page for current directory
+        with open("dir_title.html", 'w') as os_html:
+            os_html.write(header + dir_title)
+        
+        group.append(HTML('dir_title.html').render())
+
+        #Creating a list of all md files in the current directory
+        for temp in glob.glob(os.path.join(loc, operating_sys, '*.md')):
+            allmd.append(temp)
+
+        #Sorting all filenames in the directory, to maintain the order of the PDF
+        allmd.sort()
+
+        #Conversion of md to HTML
+        for md in allmd:
+
+            with open(md, "r") as inp:
+                text = inp.readlines()
+
+            with open("htmlout.html", "w") as out:
+                out.write(header)
+
+                for line in text:
+                    if re.match(r'^>', line):
+                        line = line[:0] + '####' + line[1:]
+                    html = markdown.markdown(line)
+                    out.write(html)
+                out.write(footer)
+
+            group.append(HTML('htmlout.html').render())
+            print("Rendered page {} of the directory {}".format(str(i), operating_sys))
+            i += 1
+        
+        allmd.clear()
+
+    #Merging all the documents into a single PDF
+    for doc in group:
+        for p in doc.pages:
+            ap.append(p)
+
+    #Writing the PDF to disk, preserving metadata of first tldr page
+    group[2].copy(ap).write_pdf('tldr.pdf')
+
+    if os.path.exists("tldr.pdf"):
+        print("\nCreated tldr.pdf in the current directory!\n")
+
+    #Removing unnecessary intermediate files
+    try:
+        os.remove("htmlout.html")
+        os.remove("title.html")
+        os.remove("dir_title.html")
+    except OSError:
+        print("Error removing temporary file(s)")
+
+
+if __name__ == '__main__':
+
+    #Unless specified otherwise by the user, this is the default colorscheme
+    colorscheme = "basic"
+
+    #Parsing the arguments
+    parser = argparse.ArgumentParser()
+    parser.add_argument("dir_path", help = "Path to tldr 'pages' directory")
+    parser.add_argument("-c", choices=["solarized-light", "solarized-dark"], help="Color scheme of the PDF")
+    args = parser.parse_args()
+    
+    loc = args.dir_path
+    if args.c == "solarized-light" or args.c == "solarized-dark":
+        colorscheme = args.c
+        
+    main(loc, colorscheme)
--- a/scripts/pdf/solarized-dark.css
+++ b/scripts/pdf/solarized-dark.css
@ -0,0 +1,40 @@
+@font-face {
+    font-family: 'PT_Serif-Web-Regular';
+    src: url('PT_Serif-Web-Regular.ttf')  format('truetype');
+}
+
+p {
+    margin-left: 2.5em;
+}
+
+code {
+    color: #b58900;
+}
+
+h1, h2, h4, ul {
+    font-family: "PT_Serif-Web-Regular";
+    color: #93a1a1;
+}
+
+body {
+    background-color: #002b36;
+}
+
+
+h1.titlemain {
+    text-align: center;
+    margin-top: 6em;
+    font-size: 350%;
+}
+
+h4.titlesub {
+    text-align: center;
+    font-size: 120%;
+    color: #b58900;
+}
+
+h2.titledir {
+    text-align: center;
+    margin-top: 8.2em;
+    font-size: 300%;
+}
--- a/scripts/pdf/solarized-light.css
+++ b/scripts/pdf/solarized-light.css
@ -0,0 +1,39 @@
+@font-face {
+    font-family: 'PT_Serif-Web-Regular';
+    src: url('PT_Serif-Web-Regular.ttf')  format('truetype');
+}
+
+p {
+    margin-left: 2.5em;
+}
+
+code {
+    color: #dc322f
+}
+
+h1, h2, h4, ul {
+    font-family: "PT_Serif-Web-Regular";
+    color: #586e75;
+}
+
+body {
+    background-color: #fdf6e3;
+}
+
+h1.titlemain {
+    text-align: center;
+    margin-top: 6em;
+    font-size: 350%;
+}
+
+h4.titlesub {
+    text-align: center;
+    font-size: 120%;
+    color: #dc322f
+}
+
+h2.titledir {
+    text-align: center;
+    margin-top: 8.2em;
+    font-size: 300%;
+}