Commit: 13b586d87591ed5688e94d80c44166ef20d049a0 - sword-biblia-cee (hg)

Commit MetaInfo

• Revision: 13b586d87591ed5688e94d80c44166ef20d049a0 (tree)
• Zeit: 2019-12-15 00:46:56
• Autor: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@fast...>
• Commiter: Jaime Marquínez Ferrándiz

Log Message

Commit inicial

Ändern Zusammenfassung

• add: .hgignore
• add: .idea/inspectionProfiles/profiles_settings.xml
• add: .idea/misc.xml
• add: .idea/modules.xml
• add: .idea/sword-biblia-cee.iml
• add: .idea/vcs.xml
• add: Makefile
• add: cee.conf
• add: construir_biblia.py
• add: requirements.txt
• add: template_osis.xml

Diff

diff -r 000000000000 -r 13b586d87591 .hgignore
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.hgignore	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,6 @@
+venv
+build
+# Default ignored files
+.idea/workspace.xml
+
+cee.xml

diff -r 000000000000 -r 13b586d87591 .idea/inspectionProfiles/profiles_settings.xml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.idea/inspectionProfiles/profiles_settings.xml	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 .idea/misc.xml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.idea/misc.xml	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.7 (sword-biblia-cee)" project-jdk-type="Python SDK" />
+</project>
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 .idea/modules.xml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.idea/modules.xml	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectModuleManager">
+    <modules>
+      <module fileurl="file://$PROJECT_DIR$/.idea/sword-biblia-cee.iml" filepath="$PROJECT_DIR$/.idea/sword-biblia-cee.iml" />
+    </modules>
+  </component>
+</project>
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 .idea/sword-biblia-cee.iml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.idea/sword-biblia-cee.iml	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,13 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$">
+      <excludeFolder url="file://$MODULE_DIR$/venv" />
+    </content>
+    <orderEntry type="jdk" jdkName="Python 3.7 (sword-biblia-cee)" jdkType="Python SDK" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+  <component name="TestRunnerService">
+    <option name="PROJECT_TEST_RUNNER" value="Unittests" />
+  </component>
+</module>
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 .idea/vcs.xml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.idea/vcs.xml	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="$PROJECT_DIR$" vcs="hg4idea" />
+  </component>
+</project>
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 Makefile
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/Makefile	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,22 @@
+NOMBRE_MODULO = cee
+
+OUT_DIR = build
+
+OSIS_FILE = $(NOMBRE_MODULO).xml
+
+MOD_FILES = nt nt.vss ot ot.vss
+MOD_FILES != for f in $(MOD_FILES); do echo $(OUT_DIR)/$$f; done
+ZMOD_FILES = nt.bzs nt.bzv nt.bzz ot.bzs ot.bzv ot.bzz
+ZMOD_FILES != for f in $(ZMOD_FILES); do echo $(OUT_DIR)/$$f; done
+
+all: $(MOD_FILES)
+
+$(MOD_FILES): $(OSIS_FILE)
+	mkdir -p $(OUT_DIR)
+	osis2mod $(OUT_DIR) $^
+
+$(ZMOD_FILES): $(MOD_FILES)
+	mod2zmod $(NOMBRE_MODULO) . 4 2
+
+clean:
+	rm -f $(MOD_FILES) $(ZMOD_FILES)

diff -r 000000000000 -r 13b586d87591 cee.conf
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/cee.conf	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,9 @@
+[CEE]
+ModDrv=RawText
+DataPath=./modules/texts/ztext/cee/
+CompressType=ZIP
+BlockType=BOOK
+Encoding=UTF-8
+SourceType=OSIS
+Lang=es
+Description=Biblia de la Conferencia Episcopal Española
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 construir_biblia.py
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/construir_biblia.py	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,56 @@
+import requests
+from bs4 import BeautifulSoup
+from lxml import etree as et
+
+
+NS_DEFAULT = 'http://www.bibletechnologies.net/2003/OSIS/namespace'
+NAMESPACE_MAP = {
+    '': NS_DEFAULT,
+    None: NS_DEFAULT
+}
+
+
+def osis_id_para_libro(nombre_libro: str):
+    if nombre_libro == 'Génesis':
+        return 'Gen'
+
+
+def construir_libro(url: str, nombre_libro: str, session: requests.Session):
+    print(f'Construyendo "{nombre_libro}"')
+    osis_id = osis_id_para_libro(nombre_libro)
+    r = session.get(url)
+    html_doc = r.text
+    soup = BeautifulSoup(html_doc, 'lxml')
+
+    book = et.Element('div', type='book', osisID=osis_id)
+    for chapter_tag in soup.find_all(class_='capitulo'):
+        chap_num = chapter_tag.find(class_='numcap').text
+        chapter = et.SubElement(book, 'chapter', osisID=osis_id + f'.{chap_num}')
+        for numverse_tag in chapter_tag.find_all(class_='numvers'):
+            verse = et.SubElement(chapter, 'verse', osisID=osis_id + f'.{chap_num}.{numverse_tag.text}')
+            verse.text = numverse_tag.find_next_sibling(class_='contenido').text
+    return book
+
+
+def construir_sesion():
+    """
+    Construye un objeto de sesión
+    """
+    session = requests.Session()
+    session.headers.update({'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:70.0) Gecko/20100101 Firefox/70.0'})
+    return session
+
+
+def construir_biblia():
+    parser = et.XMLParser(remove_blank_text=True)
+    osis_doc: et.ElementTree = et.parse('template_osis.xml', parser)
+    sesion = construir_sesion()
+    libro = construir_libro('https://www.conferenciaepiscopal.es/biblia/genesis', 'Génesis', sesion)
+    osis_text = osis_doc.find('osisText', NAMESPACE_MAP)
+    osis_text.append(libro)
+    with open('cee.xml', 'wb') as out_file:
+        out_file.write(et.tostring(osis_doc, pretty_print=True))
+
+
+if __name__ == '__main__':
+    construir_biblia()
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 requirements.txt
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/requirements.txt	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,3 @@
+requests
+beautifulsoup4
+lxml
\ No newline at end of file

diff -r 000000000000 -r 13b586d87591 template_osis.xml
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/template_osis.xml	Sat Dec 14 16:46:56 2019 +0100
@@ -0,0 +1,29 @@
+<?xml version='1.0' encoding='UTF-8'?>
+<osis xsi:schemaLocation='http://www.bibletechnologies.net/2003/OSIS/namespace http://www.bibletechnologies.net/osisCore.2.1.1.xsd' xmlns='http://www.bibletechnologies.net/2003/OSIS/namespace' xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance'>
+  <osisText osisIDWork='cee' xml:lang='en' osisRefWork='Bible'>
+    <header>
+      <revisionDesc>
+        <date>2019-11-16</date>
+        <p>initial OSIS 2.1.1 version</p>
+      </revisionDesc>
+      <work osisWork='cee'>
+        <title>Versión de la Conferencia Episcopal</title>
+        <contributor/>
+        <creator/>
+        <subject/>
+        <date>2016</date>
+        <description />
+        <publisher />
+        <type type='OSIS'>Bible</type>
+        <identifier type='OSIS'>cee</identifier>
+        <source>https://www.conferenciaepiscopal.es/Biblia/</source>
+        <language type='IETF'>es</language>
+        <relation/>
+        <coverage />
+        <rights>Conferencia Episcopal Española. 2016</rights>
+        <scope/>
+        <refSystem>Bible</refSystem>
+      </work>
+    </header>
+  </osisText>
+</osis>
\ No newline at end of file