Fix #14

config.yml

 site_url: http://calcul.math.cnrs.fr
-authors: Groupe Calcul
+default_author: Webmaster
 attachments_prefix: attachments/spip/
 categories:
     journees: 4
@@ -13,3 +13,5 @@ categories:
 rubriques: spip_rubriques_clean.yml
 articles: spip_articles_clean.yml
 documents: spip_documents.yml
+authors: spip_auteurs_clean.yml
+authors_links: spip_auteurs_liens.yml

spip2pelican.py

@@ -32,7 +32,7 @@ fh.setLevel(logging.INFO)
 # create console handler with higher log level and colored output
 ch = logging.StreamHandler(sys.stdout)
 ch.setLevel(logging.DEBUG)
-LOGFORMAT = "  %(log_color)s%(message)s%(reset)s"
+LOGFORMAT = "%(log_color)s%(message)s%(reset)s"
 color_formatter = ColoredFormatter(LOGFORMAT)
 ch.setFormatter(color_formatter)
 
@@ -41,8 +41,6 @@ logger.addHandler(fh)
 logger.addHandler(ch)
 
 
-# TODO: handle mix of italic and bold : { {{Marc Poinot}} (ONERA)}
-
 def header(s):
     """
     SPIP: {{{...}}}
@@ -188,8 +186,8 @@ def link(s, website):
                     art_url = re.match(r"\Aart([0-9]+)", url)
                     if art_url:
                         # [text->art#]
+                        id_art = int(art_url.group(1))
                         try:
-                            id_art = int(art_url.group(1))
                             new_url = "{filename}/" + website.article_index[id_art]
                         except KeyError:
                             new_url = nullify_url("non existing article", id_art, text, url)
@@ -334,6 +332,11 @@ class Article:
             self.summary = spip_article['descriptif']
             self.text = spip_article['texte']
 
+        try:
+            self.authors = self.website.author_index[self.type][self.id]
+        except KeyError:
+            self.authors = self.website.default_author
+
     def export_to_pelican(self):
         """
         Content of a markdown article should look like:
@@ -357,8 +360,6 @@ class Article:
             self.title = spip_to_markdown(self.title, self.website).strip()  # strip to remove any CR at end of string
 
             tags = []
-            authors = self.website.authors
-            content = spip_to_markdown(self.text, self.website)
             header = f"""\
 title: {self.title}
 date: {self.date}
@@ -366,10 +367,11 @@ modified: {self.modified}
 category: {self.category}
 tags: {tags}
 slug: {self.mdprefix}
-authors: {authors}
+authors: {self.authors}
 summary: {self.summary}
 
 """
+            content = spip_to_markdown(self.text, self.website)
             markdown = header + content
 
             export_path = os.path.join("content", self.mdpath)
@@ -388,22 +390,33 @@ class Website:
         def __missing__(self, key):
             return 0
 
+    @staticmethod
+    def _load_and_clean_yaml(filename):
+        """Load yaml file filename, clean it and return a dictionary"""
+        with open(filename, mode='r') as yml_file:
+            return yaml.load(remove_null_date(strip_invalid(yml_file)))
+
     def __init__(self, reset_output_dir=True):
         self.category_index = {}
         self.article_index = {}
         self.doc_index = {}
+
+        self.author_index = {}
         self.nullified_urls = self.MissingKeyDict()
+        self.articles = []
 
         config_filename = "config.yml"
         with open(config_filename, 'r') as ymlfile:
             cfg = yaml.load(ymlfile)
 
         self.site_url = cfg['site_url']
-        self.authors = cfg['authors']
+        self.default_author = cfg['default_author']
         self.attachments_prefix = cfg['attachments_prefix']
         self.rubriques_filename = cfg['rubriques']
         self.documents_filename = cfg['documents']
         self.articles_filename = cfg['articles']
+        self.authors_filename = cfg['authors']
+        self.authors_links_filename = cfg['authors_links']
         self.categories = {-1: "spip_divers"}
         for pelican_category, spip_rubrique in cfg['categories'].items():
             if type(spip_rubrique) == int:
@@ -441,8 +454,8 @@ class Website:
             return self.categories[id_rubrique]
 
         # Load original rubriques file as a list
-        with open(self.rubriques_filename, mode='r') as yml_rubriques:
-            rubriques = yaml.load(yml_rubriques.read())
+        with open(self.rubriques_filename, mode='r') as yml_file:
+            rubriques = yaml.load(yml_file.read())
 
         parents = {rubrique['id_rubrique']: rubrique['id_parent'] for rubrique in rubriques}
         self.category_index = {rubrique['id_rubrique']: get_category(rubrique['id_rubrique'])
@@ -452,10 +465,27 @@ class Website:
         """Build the index dictionary: {id_doc: file_path}"""
 
         # Load original document file as a list
-        with open(self.documents_filename, mode='r') as yml_doc:
-            docs = yaml.load(remove_null_date(strip_invalid(yml_doc)))
+        docs = self._load_and_clean_yaml(self.documents_filename)
         self.doc_index = {doc['id_document']: doc['fichier'] for doc in docs}
 
+    def _build_author_index(self):
+        """Build the index dictionary: {spip_type: art_id: author_name}"""
+
+        # Load author file as a list
+        authors = self._load_and_clean_yaml(self.authors_filename)
+        author_name_index = {author['id_auteur']: author['nom'] for author in authors}
+
+        # Load article/author file as a list
+        authors_links = self._load_and_clean_yaml(self.authors_links_filename)
+
+        for authors_link in authors_links:
+            spip_type = authors_link['objet']
+            art_id = authors_link['id_objet']
+            author_id = authors_link['id_auteur']
+            if spip_type not in self.author_index.keys():
+                self.author_index[spip_type] = {}
+            self.author_index[spip_type][art_id] = author_name_index[author_id]
+
     def _build_articles(self):
         """
         Build:
@@ -473,20 +503,23 @@ class Website:
                     if not article.skip_reason:
                         self.article_index[article.id] = article.mdpath
 
-        self.articles = []
-        self.article_index = {}
         add_articles(self.articles_filename, 'article')
         add_articles(self.rubriques_filename, 'rubrique')
 
     def read_spip(self):
         """Read spip yaml files to build useful indices and article list"""
+        logger.debug("-------")
+        logger.debug("Loading Spip data")
         self._build_category_index()
         self._build_doc_index()
+        self._build_author_index()
         self._build_articles()
 
     def export_to_pelican(self):
         """Loop on Spip articles to convert them into Pelican format"""
 
+        logger.debug("-------")
+        logger.debug("Exporting to Pelican")
         processed = []
         for article in self.articles:
             skip_reason = article.export_to_pelican()

yaml_cleaner.py

@@ -27,7 +27,7 @@ def remove_null_date(s):
     """
     Remove "date:",  "date_tmp:", etc. if equals to 0000-00-00 00:00:00 (otherwise yaml.load() would fail)
     """
-    s = re.sub(r'^  date.*: 0000-00-00 00:00:00$', r'', s, flags=re.MULTILINE)
+    s = re.sub(r'(^  date|^  en_ligne).*: 0000-00-00 00:00:00$', r'', s, flags=re.MULTILINE)
     return s
 
 
@@ -37,12 +37,14 @@ def clean_titles(s):
     """
     def title_replace(matchobj):
         """Return a clean title line"""
-        title = matchobj.group(1)
+        title_type = matchobj.group(1)
+        title = matchobj.group(2)
         title = re.sub(r"'", "''", title)  # Avoid real text simple quote to be interpreted as end of string
-        return f"  titre: '{title}'"
+        return f"{title_type}'{title}'"
 
-    res = re.sub('^  titre: (.*)$', title_replace, s, flags=re.MULTILINE)
-    return res
+    s = re.sub('^(  titre: )(.*)$', title_replace, s, flags=re.MULTILINE)
+    s = re.sub('^(  nom_site: )(.*)$', title_replace, s, flags=re.MULTILINE)
+    return s
 
 
 def force_encode(line, iline, codecs=('cp1252', 'utf8')):
@@ -91,5 +93,6 @@ def clean_yaml(yml_filename):
 
 if __name__ == '__main__':
 
+    clean_yaml("spip_auteurs.yml")
     clean_yaml("spip_articles.yml")
     clean_yaml("spip_rubriques.yml")