Auto-generate labels from a base libreoffice template

8 files changed, 192 insertions, 3 deletions

diff --git a/archaeological_finds/tests.py b/archaeological_finds/tests.py
index 548a74e7b..93548f8fc 100644
--- a/archaeological_finds/tests.py
+++ b/archaeological_finds/tests.py
@@ -38,7 +38,8 @@ from django.core.urlresolvers import reverse
 from django.test import tag
 from django.test.client import Client
 from ishtar_common.models import ImporterType, IshtarUser, ImporterColumn,\
-    FormaterType, ImportTarget, IshtarSiteProfile, ProfileType
+    FormaterType, ImportTarget, IshtarSiteProfile, ProfileType, ImporterModel, \
+    DocumentTemplate
 from django.utils.text import slugify
 from django.utils.translation import pgettext_lazy, gettext_lazy as _
 
@@ -2513,3 +2514,68 @@ class PublicAPITest(FindInit, APITestCase):
                 value = value[key]
             self.assertEqual(value, result)
 
+
+class LabelTest(FindInit, TestCase):
+    fixtures = FIND_FIXTURES
+    model = models.Find
+
+    def setUp(self):
+        templates = [
+            settings.ROOT_PATH + '../archaeological_finds/tests/' + t
+            for t in ("etiquettes-mobilier.odt", "etiquettes-mobilier",
+                      "etiquettes-mobilier.txt", "bad_lo.zip",
+                      "truncated_xml.zip")
+        ]
+        self.templates = []
+        for template in templates:
+            filename = template.split("/")[-1]
+            shutil.copy(template,
+                        os.path.join(settings.MEDIA_ROOT, filename),
+                        follow_symlinks=True)
+            self.templates.append(
+                os.path.join(settings.MEDIA_ROOT, filename))
+
+    def tearDown(self):
+        for tpl in self.templates:
+            if os.path.exists(tpl):
+                os.remove(tpl)
+
+    def test_label(self):
+        base_targets = ";".join("Cadre{}".format(idx) for idx in range(1, 25))
+        base_tpl, missing_ext, text_file, bad_lo, trunc_xml = self.templates
+        dataset = (
+            (base_tpl, base_targets, True, "OK"),
+            (base_tpl, "", False, "no target"),
+            (base_tpl, "-;Cadre2;Cadre3", False, "bad first target"),
+            (base_tpl, "Cadre1;Frame2;Frame3", True,
+             "first target OK, silently failed other targets"),
+            (missing_ext, base_targets, True, "missing extension"),
+            (text_file, base_targets, False, "text file"),
+            (bad_lo, base_targets, False, "missing content.xml"),
+            (trunc_xml, base_targets, False, "truncated content.xml"),
+        )
+        for tpl_file, targets, is_ok, msg in dataset:
+            with open(tpl_file, 'rb') as tpl:
+                template = SimpleUploadedFile("etiquettes-mobilier.odt",
+                                              tpl.read())
+                model, __ = ImporterModel.objects.get_or_create(
+                    klass='archaeological_finds.models.Find'
+                )
+                q = DocumentTemplate.objects.filter(slug="test")
+                if q.count():
+                    q.all()[0].delete()
+                doc = DocumentTemplate.objects.create(
+                    name="Test",
+                    slug="test",
+                    associated_model=model,
+                    available=True,
+                    label_targets=targets,
+                    label_template=template)
+            self.templates.append(doc.label_template.path)
+            doc = DocumentTemplate.objects.get(pk=doc.pk)
+            msg = "Fail on dataset: " + msg
+            if is_ok:
+                self.assertTrue(doc.template.name, msg=msg)
+                self.templates.append(doc.template.path)
+            else:
+                self.assertFalse(doc.template.name, msg=msg)
diff --git a/archaeological_finds/tests/bad_lo.zip b/archaeological_finds/tests/bad_lo.zip
new file mode 100644
index 000000000..f05cd2d0b
--- /dev/null
+++ b/archaeological_finds/tests/bad_lo.zip
diff --git a/archaeological_finds/tests/etiquettes-mobilier b/archaeological_finds/tests/etiquettes-mobilier
new file mode 120000
index 000000000..89ec4030b
--- /dev/null
+++ b/archaeological_finds/tests/etiquettes-mobilier
@@ -0,0 +1 @@
+etiquettes-mobilier.odt
\ No newline at end of file
diff --git a/archaeological_finds/tests/etiquettes-mobilier.odt b/archaeological_finds/tests/etiquettes-mobilier.odt
new file mode 100644
index 000000000..7a8ffdedb
--- /dev/null
+++ b/archaeological_finds/tests/etiquettes-mobilier.odt
diff --git a/archaeological_finds/tests/etiquettes-mobilier.txt b/archaeological_finds/tests/etiquettes-mobilier.txt
new file mode 100644
index 000000000..de2d58990
--- /dev/null
+++ b/archaeological_finds/tests/etiquettes-mobilier.txt
@@ -0,0 +1 @@
+Non libreoffice file
diff --git a/archaeological_finds/tests/truncated_xml.zip b/archaeological_finds/tests/truncated_xml.zip
new file mode 100644
index 000000000..830918c5a
--- /dev/null
+++ b/archaeological_finds/tests/truncated_xml.zip
diff --git a/ishtar_common/migrations/0210_auto_20210106_1127.py b/ishtar_common/migrations/0210_auto_20210106_1127.py
new file mode 100644
index 000000000..3dcf7e7ce
--- /dev/null
+++ b/ishtar_common/migrations/0210_auto_20210106_1127.py
@@ -0,0 +1,30 @@
+# -*- coding: utf-8 -*-
+# Generated by Django 1.11.27 on 2021-01-06 11:27
+from __future__ import unicode_literals
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('ishtar_common', '0209_auto_20210105_1712'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='documenttemplate',
+            name='label_targets',
+            field=models.TextField(blank=True, help_text='Each target is separated by a semi-colon. The first target is the name of the object including the data in base template. Following targets will be filled with the content of the first target. For instance: "Cadre1;Cadre2;Cadre3;Cadre4;Cadre5;Cadre6" for a sheet with 6 labels.', null=True, verbose_name='Labels: targets for labels in the LibreOffice file'),
+        ),
+        migrations.AddField(
+            model_name='documenttemplate',
+            name='label_template',
+            field=models.FileField(blank=True, help_text='La taille maximale supportée pour le fichier est de 100 Mo.', null=True, upload_to='templates/%Y/', verbose_name='Base template for labels'),
+        ),
+        migrations.AlterField(
+            model_name='documenttemplate',
+            name='template',
+            field=models.FileField(blank=True, help_text='La taille maximale supportée pour le fichier est de 100 Mo.', null=True, upload_to='templates/%Y/', verbose_name='Template'),
+        ),
+    ]
diff --git a/ishtar_common/models.py b/ishtar_common/models.py
index 23416d406..6ac197af4 100644
--- a/ishtar_common/models.py
+++ b/ishtar_common/models.py
@@ -20,6 +20,7 @@
 """
 Models description
 """
+from bs4 import BeautifulSoup
 import copy
 import datetime
 import inspect
@@ -55,6 +56,7 @@ from django.contrib.sites.models import Site
 from django.core.cache import cache
 from django.core.exceptions import ObjectDoesNotExist, ValidationError, \
     MultipleObjectsReturned
+from django.core.files.base import ContentFile
 from django.core.files.uploadedfile import SimpleUploadedFile
 from django.core.urlresolvers import reverse
 from django.db.models import Q, Max, Count
@@ -1436,7 +1438,21 @@ class DocumentTemplate(models.Model):
     slug = models.SlugField(_("Slug"), max_length=100, unique=True)
     associated_model = models.ForeignKey(ImporterModel)
     template = models.FileField(
-        _("Template"), upload_to="templates/%Y/", help_text=max_size_help())
+        _("Template"), upload_to="templates/%Y/", blank=True, null=True,
+        help_text=max_size_help())
+    label_template = models.FileField(
+        _("Base template for labels"), upload_to="templates/%Y/",
+        blank=True, null=True, help_text=max_size_help())
+    label_targets = models.TextField(
+        _("Labels: targets for labels in the LibreOffice file"),
+        blank=True, null=True,
+        help_text=_("Each target is separated by a semi-colon. The first "
+                    "target is the name of the object including the data in "
+                    "base template. Following targets will be filled with the "
+                    "content of the first target. For instance: "
+                    "\"Cadre1;Cadre2;Cadre3;Cadre4;Cadre5;Cadre6\" for a "
+                    "sheet with 6 labels.")
+    )
     available = models.BooleanField(_("Available"), default=True)
     for_labels = models.BooleanField(_("Used for labels"), default=False)
     label_per_page = models.IntegerField(
@@ -1462,10 +1478,85 @@ class DocumentTemplate(models.Model):
             raise ValidationError(_("For label template, you must provide "
                                     "number of label per page."))
 
+    def generate_label_template(self):
+        if not self.label_template.name or not self.label_targets:
+            return
+        targets = self.label_targets.split(";")
+        base_target = targets[0]
+        try:
+            with zipfile.ZipFile(self.label_template.path) as zip:
+                with zip.open('content.xml') as content:
+                    soup = BeautifulSoup(content.read(), 'xml')
+                    base_content = soup.find(
+                        "draw:frame", attrs={"draw:name": base_target})
+                    if not base_content:
+                        return
+                    base_content = base_content.contents
+        except (FileNotFoundError, zipfile.BadZipFile, KeyError):
+            base_content = None
+        if not base_content:
+            return
+        for idx, target in enumerate(targets[1:]):
+            replace_str = "items." + str(idx + 1)
+            new_content = []
+            for content in base_content:
+                content = copy.copy(content)
+                for text in content.find_all(text=re.compile("items.0")):
+                    fixed_text = text.replace("items.0", replace_str)
+                    text.replace_with(fixed_text)
+                for image in content.find_all(
+                        attrs={"draw:name": re.compile("items.0")}):
+                    image["draw:name"] = image["draw:name"].replace("items.0",
+                                                                    replace_str)
+                new_content.append(content)
+            next_target = soup.find(
+                    "draw:frame", attrs={"draw:name": target})
+            if next_target:
+                next_target.contents = new_content
+
+        with tempfile.TemporaryDirectory() as tmp:
+            sp = self.label_template.name.split(os.sep)[-1].split(".")
+            if len(sp) == 1:  # no extension?
+                sp.append("odt")
+            sp[-2] += "-label"
+            new_filename = ".".join(sp)
+            new_file = os.path.join(tmp, new_filename)
+            with zipfile.ZipFile(new_file, 'w') as zip_out:
+                with zipfile.ZipFile(self.label_template.path, 'r') as zip_in:
+                    zip_out.comment = zip_in.comment
+                    for item in zip_in.infolist():
+                        if item.filename != "content.xml":
+                            zip_out.writestr(item,
+                                             zip_in.read(item.filename))
+            with zipfile.ZipFile(new_file, mode='a',
+                                 compression=zipfile.ZIP_DEFLATED) as zf:
+                zf.writestr("content.xml", str(soup))
+
+            media_dir = "templates/{}/".format(datetime.date.today().year)
+            full_media_dir = os.path.join(settings.MEDIA_ROOT, media_dir)
+            if not os.path.exists(full_media_dir):
+                os.mkdir(full_media_dir)
+            media_file = new_filename
+            idx = 0
+            while os.path.exists(os.path.join(settings.MEDIA_ROOT, media_file)):
+                idx += 1
+                sp = media_file.split(".")
+                sub_sp = sp[-2].split("-label")
+                sub_sp[-1] += str(idx)
+                sp[-2] = "-label".join(sub_sp)
+                media_file = ".".join(sp)
+            with open(new_file, "rb") as file:
+                with ContentFile(file.read()) as file_content:
+                    self.template.save(media_file, file_content)
+                    self.save()
+
     def save(self, *args, **kwargs):
         if not self.slug:
             self.slug = create_slug(DocumentTemplate, self.name)
-        return super(DocumentTemplate, self).save(*args, **kwargs)
+        super(DocumentTemplate, self).save(*args, **kwargs)
+        if self.label_template.name and self.label_targets and not \
+                self.template:
+            self.generate_label_template()
 
     @classmethod
     def get_tuples(cls, dct=None, empty_first=True):