Thumbnails - Restore integration

.gitignore

@@ -142,3 +142,4 @@ junit/*
 !tests/fixtures/template_Isogeo.docx
 
 tests/fixtures/api_search_*.json
+tests/fixtures/thumbnails.csv

isogeotodocx/isogeo2docx.py

@@ -16,7 +16,7 @@
 
 # 3rd party library
 from docxtpl import DocxTemplate, InlineImage, RichText, etree
-from isogeo_pysdk import Event, Isogeo, IsogeoTranslator, IsogeoUtils, Metadata, Share
+from isogeo_pysdk import Event, IsogeoTranslator, IsogeoUtils, Metadata, Share
 
 # custom submodules
 from isogeotodocx.utils import Formatter
@@ -37,6 +37,7 @@ class Isogeo2docx(object):
     """IsogeoToDocx class.
 
     :param str lang: selected language for output
+    :param dict thumbnails: dictionary of metadatas associated to an image path
     :param str url_base_edit: base url to format edit links (basically app.isogeo.com)
     :param str url_base_view: base url to format view links (basically open.isogeo.com)
     """
@@ -45,20 +46,11 @@ def __init__(
         self,
         lang="FR",
         default_values=("NR", "1970-01-01T00:00:00+00:00"),
+        thumbnails: dict = None,
         url_base_edit: str = "https://app.Isogeo.com",
         url_base_view: str = "https://open.isogeo.com",
     ):
-        """Common variables for Word processing.
-
-        default_values (optional) -- values used to replace missing values.
-        Must be a tuple with 2 values structure:
-        (
-        str_for_missing_strings_and_integers,
-        str_for_missing_dates
-        )
-
-
-        """
+        """Processing matching between Isogeo metadata and a Miscrosoft Word template."""
         super(Isogeo2docx, self).__init__()
 
         # ------------ VARIABLES ---------------------
@@ -91,6 +83,13 @@ def __init__(
         # FORMATTER
         self.fmt = Formatter(output_type="Word")
 
+        # THUMBNAILS
+        if thumbnails is not None and isinstance(thumbnails, dict):
+            self.thumbnails = thumbnails
+        else:
+            self.thumbnails = {}
+            logger.debug("No valid thumbnails matching table passed.")
+
         # URLS
         utils.app_url = url_base_edit  # APP
         utils.oc_url = url_base_view  # OpenCatalog url
@@ -282,7 +281,6 @@ def md2docx(self, docx_template: DocxTemplate, md: Metadata, share: Share = None
 
         # FILLFULLING THE TEMPLATE #
         context = {
-            # "varThumbnail": InlineImage(docx_template, md.thumbnail),
             "varTitle": self.fmt.clean_xml(md.title),
             "varAbstract": self.fmt.clean_xml(md.abstract),
             "varNameTech": md.name,
@@ -323,6 +321,14 @@ def md2docx(self, docx_template: DocxTemplate, md: Metadata, share: Share = None
             "varEditAPP": link_edit,
         }
 
+        # -- THUMBNAIL -----------------------------------------------------------------
+        if md._id in self.thumbnails and Path(self.thumbnails.get(md._id)).is_file():
+            thumbnail = str(Path(self.thumbnails.get(md._id)).resolve())
+            context["varThumbnail"] = InlineImage(docx_template, thumbnail)
+            logger.info(
+                "Thumbnail found for {}: {}".format(md.title_or_name(1), thumbnail)
+            )
+
         # fillfull file
         try:
             docx_template.render(context, autoescape=True)
@@ -366,9 +372,10 @@ def missing_values(self, idx_type=0):
 # ###################################
 if __name__ == "__main__":
     """
-        Standalone execution and tests
+        Standalone execution and basic tests
     """
     # ------------ Specific imports ----------------
+    from csv import DictReader
     from dotenv import load_dotenv
     from logging.handlers import RotatingFileHandler
     from os import environ
@@ -406,6 +413,10 @@ def missing_values(self, idx_type=0):
     # get user ID as environment variables
     load_dotenv("dev.env")
 
+    # misc
+    METADATA_TEST_FIXTURE_UUID = environ.get("ISOGEO_FIXTURES_METADATA_COMPLETE")
+    WORKGROUP_TEST_FIXTURE_UUID = environ.get("ISOGEO_WORKGROUP_TEST_UUID")
+
     # ignore warnings related to the QA self-signed cert
     if environ.get("ISOGEO_PLATFORM").lower() == "qa":
         urllib3.disable_warnings()
@@ -422,10 +433,6 @@ def missing_values(self, idx_type=0):
     # getting a token
     isogeo.connect()
 
-    # misc
-    METADATA_TEST_FIXTURE_UUID = environ.get("ISOGEO_FIXTURES_METADATA_COMPLETE")
-    WORKGROUP_TEST_FIXTURE_UUID = environ.get("ISOGEO_WORKGROUP_TEST_UUID")
-
     # ------------ Isogeo search --------------------------
     search_results = isogeo.search(
         include="all",
@@ -441,11 +448,24 @@ def missing_values(self, idx_type=0):
     Path("_output/").mkdir(exist_ok=True)
 
     # template
-    template_path = Path(r"tests\fixtures\template_Isogeo.docx")
+    template_path = Path("tests/fixtures/template_Isogeo.docx")
     assert template_path.is_file()
 
+    # thumbnails table
+    thumbnails_table_csv_path = Path("tests/fixtures/thumbnails.csv")
+    assert thumbnails_table_csv_path.is_file()
+
+    # CSV structure
+    csv_headers = ["isogeo_uuid", "isogeo_title_slugged", "img_abs_path"]
+    thumbnails_dict = {}
+    with thumbnails_table_csv_path.open("r", newline="") as csv_thumbnails:
+        reader = DictReader(csv_thumbnails, fieldnames=csv_headers)
+        next(reader, None)  # skip header line
+        for row in reader:
+            thumbnails_dict[row.get("isogeo_uuid")] = row.get("img_abs_path")
+
     # instanciate
-    toDocx = Isogeo2docx()
+    toDocx = Isogeo2docx(thumbnails=thumbnails_dict)
 
     # parse results and export it
     for md in search_results.results:

tests/fixturing.py

@@ -12,21 +12,27 @@
 # ##################################
 
 # Standard library
+import csv
 import json
 import logging
-from os import environ, path
+from os import environ
+from pathlib import Path
+from random import sample
 
 # 3rd party
 from dotenv import load_dotenv
 import urllib3
 
 # Isogeo
-from isogeo_pysdk import Isogeo
+from isogeo_pysdk import Isogeo, Metadata, MetadataSearch
 
 # #############################################################################
 # ######## Globals #################
 # ##################################
 
+# BASE DIRECTORY - Script is meant to be launched from the root of the repository
+BASE_DIR = Path(__file__).parent
+
 # env vars
 load_dotenv("dev.env", override=True)
 
@@ -49,6 +55,11 @@
 # ########## Fixturing ###############
 # ####################################
 
+# -- RETRIEVE ISOGEO SEARCH RESULTS ----------------------------------------------------
+logger.info(
+    "Connecting to Isogeo API to download search results to be used by the following test."
+)
+
 # instanciating the class
 isogeo = Isogeo(
     auth_mode="group",
@@ -59,32 +70,80 @@
 )
 isogeo.connect()
 
-# Downloading directly from Isogeo API
-BASE_DIR = path.dirname(path.abspath(__file__))
-
 # complete search - only Isogeo Tests
-out_search_complete_tests = path.join(
-    BASE_DIR, "fixtures", "api_search_complete_tests.json"
-)
-if not path.isfile(out_search_complete_tests):
+out_search_complete_tests = BASE_DIR / "fixtures" / "api_search_complete_tests.json"
+
+if not out_search_complete_tests.is_file():
     request = isogeo.search(
         query="owner:{}".format(ISOGEO_WORKGROUP_TEST_UUID),
         whole_results=1,
         include="all",
         augment=1,
     )
-    with open(out_search_complete_tests, "w") as json_basic:
+    with out_search_complete_tests.open("w") as json_basic:
         json.dump(request.to_dict(), json_basic, sort_keys=True)
 else:
-    logging.info("JSON already exists. If you want to update it, delete it first.")
+    logger.info(
+        "JSON already exists at: {}. If you want to update it, delete it first.".format(
+            out_search_complete_tests.resolve()
+        )
+    )
 
 # complete search
-out_search_complete = path.join(BASE_DIR, "fixtures", "api_search_complete.json")
-if not path.isfile(out_search_complete):
+out_search_complete = BASE_DIR / "fixtures" / "api_search_complete.json"
+
+if not out_search_complete.is_file():
     request = isogeo.search(whole_results=1, include="all", augment=1)
-    with open(
-        path.join(BASE_DIR, "fixtures", "api_search_complete.json"), "w"
-    ) as json_basic:
+    with out_search_complete.open("w") as json_basic:
         json.dump(request.to_dict(), json_basic, sort_keys=True)
 else:
-    logging.info("JSON already exists. If you want to update it, delete it first.")
+    logger.info(
+        "JSON already exists at: {}. If you want to update it, delete it first.".format(
+            out_search_complete.resolve()
+        )
+    )
+
+
+# -- BUILD THUMBNAILS TABLE ------------------------------------------------------------
+logger.info("Write the thumbnails table to test image insertion into Word export.")
+
+thumbnails_table_out = BASE_DIR / "fixtures" / "thumbnails.csv"
+
+# load previous genereated JSON
+with out_search_complete.open("r") as f:
+    search = json.loads(f.read())
+search = MetadataSearch(**search)
+
+fixtures_images = list(Path(BASE_DIR / "fixtures" / "img").iterdir())
+
+with thumbnails_table_out.open("w", newline="") as csvfile:
+    # CSV structure
+    csv_headers = ["isogeo_uuid", "isogeo_title_slugged", "img_abs_path"]
+    # write headers
+    writer = csv.DictWriter(csvfile, fieldnames=csv_headers)
+    writer.writeheader()
+
+    # pick random metadata
+    metadatas = sample(search.results, len(fixtures_images))
+
+    # parse fixtures images
+    for image, md in zip(fixtures_images, metadatas):
+        metadata = Metadata.clean_attributes(md)
+        writer.writerow(
+            {
+                "isogeo_uuid": metadata._id,
+                "isogeo_title_slugged": metadata.title_or_name(slugged=1),
+                "img_abs_path": image.resolve(),
+            }
+        )
+
+    # force a thumbnail for the fixture metadata
+    writer.writerow(
+        {
+            "isogeo_uuid": environ.get("ISOGEO_FIXTURES_METADATA_COMPLETE"),
+            "isogeo_title_slugged": "ISOGEO FIXTURES METADATA COMPLETE",
+            "img_abs_path": image.resolve(),
+        }
+    )
+
+logger.info("{} thumbnails associated with metadatas.".format(len(fixtures_images) + 1))