From 8f032d2e214a93a4cdd09ed375d64f8e212673f5 Mon Sep 17 00:00:00 2001
From: Lowik Bourgeoisat <lowik.bourgeoisat@geneanet.org>
Date: Wed, 20 Feb 2019 17:42:06 +0100
Subject: [PATCH] Add a helper method to generate TableElement code

---
 .../applications/weboobdebug/weboobdebug.py   |  5 ++-
 weboob/browser/elements.py                    | 35 +++++++++++++++++++
 2 files changed, 39 insertions(+), 1 deletion(-)

diff --git a/weboob/applications/weboobdebug/weboobdebug.py b/weboob/applications/weboobdebug/weboobdebug.py
index 587625e1d5..b8872dd898 100644
--- a/weboob/applications/weboobdebug/weboobdebug.py
+++ b/weboob/applications/weboobdebug/weboobdebug.py
@@ -22,6 +22,7 @@
 from optparse import OptionGroup
 
 from weboob.tools.application.base import Application
+from weboob.browser.elements import generate_table_element
 
 
 class WeboobDebug(Application):
@@ -57,7 +58,9 @@ def main(self, argv):
             print(u'Unable to load backend "%s"' % backend_name, file=self.stderr)
             return 1
 
-        locs = dict(backend=backend, browser=backend.browser, application=self, weboob=self.weboob)
+        locs = dict(backend=backend, browser=backend.browser,
+                    application=self, weboob=self.weboob,
+                    generate_table_element=generate_table_element)
         banner = 'Weboob debug shell\nBackend "%s" loaded.\nAvailable variables:\n' % backend_name \
                  + '\n'.join(['  %s: %s' % (k, v) for k, v in locs.items()])
 
diff --git a/weboob/browser/elements.py b/weboob/browser/elements.py
index 39df7a7d46..8f01f78b34 100644
--- a/weboob/browser/elements.py
+++ b/weboob/browser/elements.py
@@ -40,6 +40,41 @@
 __all__ = ['DataError', 'AbstractElement', 'ListElement', 'ItemElement', 'TableElement', 'SkipItem']
 
 
+def generate_table_element(doc, head_xpath, cleaner=CleanText):
+    """
+    Prints generated base code for TableElement/TableCell usage.
+    It is intended for development purposes, typically in weboob-debug.
+    :param doc: lxml tree of the page (e.g. browser.page.doc)
+    :param head_xpath: xpath of header columns (e.g. //table//th)
+    :type head_xpath: str
+    :param cleaner: cleaner class (Filter)
+    :type cleaner: Filter
+    """
+    from unidecode import unidecode
+    indent = 4
+    headers = doc.xpath(head_xpath)
+    cols = dict()
+    for el in headers:
+        th = cleaner.clean(el)
+        cols.update({re.sub('[^a-zA-Z]', '_', unidecode(th)).lower(): th})
+
+    print(' ' * indent + '@method')
+    print(' ' * indent + 'class get_items(TableElement):')
+    if cleaner is not CleanText:
+        print(' ' * indent * 2 + 'cleaner = %s' % cleaner.__name__)
+    print(' ' * indent * 2 + 'head_xpath = ' + repr(head_xpath))
+    print(' ' * indent * 2 + 'item_xpath = ' + repr('...') + '\n')
+
+    for col, name in cols.items():
+        print(' ' * indent * 2 + 'col_' + col + ' = ' + repr(name))
+
+    print('\n' + ' ' * indent * 2 + 'class item(ItemElement):')
+    print(' ' * indent * 3 + 'klass = BaseObject' + '\n')
+
+    for col in cols:
+        print(' ' * indent * 3 + 'obj_' + col + ' = ' + "TableCell('%s') & CleanText()" % col)
+
+
 class DataError(Exception):
     """
     Returned data from pages are incoherent.
-- 
GitLab