Update the setup instructions and add a title.

2 files changed, 26 insertions, 6 deletions

diff --git a/README b/README
index fda108e..53582c1 100644
--- a/README
+++ b/README
@@ -1,3 +1,6 @@
+Scrapers for norweigan post journal sources
+===========================================
+
 Classic API code available from
 
 https://bitbucket.org/ScraperWiki/scraperwiki-classic/src/c7f076950476?at=default
@@ -8,13 +11,12 @@ Standalone lib https://github.com/scraperwiki/scraperwiki-python
 
 == Running / testing scrapers ==
 
-In addition to checking out the repo, the following is required to test or
-run most scrapers:
+To get the scrapers running, one need to set up the data directory and
+a patched copy of the scraperwiki-python project.  The script
+env-setup is provided to do so.  Run it from the top of the checked
+out scraper directory to set up your own copy.
 
-mkdir data
-scp -r 'scraper.nuug.no:/srv/scraper/postjournaler/testlib/*' .
-apt-get install python-alembic python-beautifulsoup python-dateutil
-cp scrapersources/postliste-python-lib scrapersources/postliste-python-lib.py
+  ./env-setup
 
 To run a scraper, use the run-scraper command and give the scraper
 name as the argument.  For example like this:
diff --git a/env-setup b/env-setup
new file mode 100755
index 0000000..e4987f4
--- /dev/null
+++ b/env-setup
@@ -0,0 +1,18 @@
+#!/bin/sh
+
+# Set up the local scraperwiki python library
+if [ ! -d testlib ] ; then
+    mkdir testlib
+    git clone https://github.com/petterreinholdtsen/scraperwiki-python.git \
+	testlib/scraperwiki-python
+    (cd testlib/scraperwiki-python; git checkout -b localbranch)
+    (cd testlib/scraperwiki-python; git merge -m "Merge patches." origin/scraperwiki.swimport \
+	origin/sqliteerror origin/verbose-sqlite)
+fi
+
+# Install the rest
+sudo apt-get install python-alembic python-beautifulsoup python-dateutil
+if [ ! -h scrapersources/postliste-python-lib.py ] ; then
+    ln -s postliste-python-lib scrapersources/postliste-python-lib.py
+fi
+mkdir -p data