article scraping, printing and template. barcodes and db for links.

8 years ago · 44914e90e7
parent 85438784f8
commit 44914e90e7
2 changed files with 117 additions and 1 deletions
--- a/screenless/bureau/publications/article.html
+++ b/screenless/bureau/publications/article.html
@ -0,0 +1,55 @@
+<!DOCTYPE html>
+<html>
+  <title>${title}</title>
+  <meta charset="utf-8">
+  <style type="text/css">
+    body {
+      font-family: Junicode;
+      font-size: 10pt;
+      line-height: 1;
+      counter-reset: sidenote-counter;
+    }
+    h1 {
+      font-variant: small-caps;
+      width: 67%;
+    }
+    p {
+      width: 67%;
+    }
+    img {
+      width: 67%;
+    }
+    a {
+      text-decoration: none;
+      color: inherit;
+    }
+    a::after {
+      counter-increment: sidenote-counter;
+      content: counter(sidenote-counter);
+      font-size: smaller;
+      vertical-align: super;
+    }
+    #date #article-src {
+      padding-bottom: 10%;
+      font-style: italic;
+    }
+    .sidenote {
+      float: right;
+      clear: right;
+      margin-right: -30%;
+      width: 30%;
+      height: 30%;
+      position: relative;
+    }
+    .sidenote svg {
+      width: 100%;
+      height: 100%;
+    }
+  </style>
+<body>
+  <h1>${title}</h1>
+  <p id="date">${date}</p>
+  ${article}
+  <p id="article-src">Article Source: ${url}</p>
+</body>
+</html>
--- a/screenless/bureau/publications/publications.py
+++ b/screenless/bureau/publications/publications.py
@ -1,7 +1,15 @@
+# publications office for reading and writing
+from datetime import datetime
 import json
 import os
+import random
+import string
+import urllib.request, urllib.parse, urllib.error

+import code128
+import lxml.html
 import PIL
+from readability import readability

 from bureau import Bureau, add_command, add_api

@ -19,9 +27,15 @@ class Publications(Bureau):

    def __init__(self):
        Bureau.__init__(self)
+
+        # set up db for published sites
+        # TODO: rename this to something less ambiguous
        self.db = os.path.expanduser("~/.screenless/PB.data")
        if not os.path.exists(self.db):
            os.mkdir(self.db)
+
+        # set up urldb for short-codes
+        self.urldb = self.dbenv.open_db(b"urldb")
        

    @add_command("new", "Create a new Publication/Site")
@ -56,7 +70,54 @@ class Publications(Bureau):
    def _update_page(self, site, page):
        pass

-
+    @add_command("r", "Print a web page for reading")
+    def print_url(self, data):
+        """
+        Print out a web page for reading.  The command requires a short-code,
+        typically referenced via barcode.  Short-codes refer to full resource
+        URIs recorded in the Publications office 'urldb' database.
+        """
+        shortcode, _ = data.split(".")
+        with self.dbenv.begin(db=self.urldb) as txn:
+            print("looking up short-code:", shortcode)
+            url = txn.get(shortcode.encode('utf-8')).decode()
+
+        if not url:
+            print("ERROR: no valid URL in db for short code: ", shortcode)
+            return
+
+        # download 
+        headers = {'User-Agent': 'Mozilla/5.0'}
+        req = urllib.request.Request(url, None, headers)
+        urldata = urllib.request.urlopen(req)
+
+        # re-render with readability
+        doc = readability.Document(urldata.read(),
+                                   url=url)
+        timestamp = datetime.now().strftime("Sourced %d %B, %Y at %H:%M")
+       
+        html = lxml.html.document_fromstring(doc.summary())
+
+        # store links then make corresponding svg barcodes
+        for link in html.findall(".//a"):
+            tmpcode = ''.join(random.choice(string.ascii_letters +\
+                                            string.digits)\
+                              for _ in range(5))
+
+            with self.dbenv.begin(write=True, db=self.urldb) as txn:
+                if "href" in link.attrib:
+                    txn.put(tmpcode.encode(), link.attrib["href"].encode())
+
+            svg = code128.svg("PBr." + tmpcode)
+
+            sidenote = html.makeelement("span")
+            sidenote.attrib["class"] = "sidenote"
+            sidenote.append(lxml.html.fromstring(svg.encode()))
+            link.addnext(sidenote)
+
+        self.print_full("article.html", title=doc.title(),
+                        article=lxml.html.tostring(html).decode("utf-8"),
+                        url=url, date=timestamp)


 def main():