From 73fe8ac3187ce70e85356fffb30ddfd35ad857e7 Mon Sep 17 00:00:00 2001
From: Joshua Hampton <joshua.hampton@live.co.uk>
Date: Fri, 1 Dec 2023 12:52:28 +0000
Subject: [PATCH] Auto-find specs for NCAS-IMAGE files

---
 checksit/check.py       | 43 +++++++++++++++++++++++++++++++++--------
 checksit/readers/cdl.py |  2 +-
 2 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/checksit/check.py b/checksit/check.py
index 32473226..af5b3f80 100644
--- a/checksit/check.py
+++ b/checksit/check.py
@@ -216,12 +216,16 @@ def check_file(self, file_path, template="auto", mappings=None, extra_rules=None
 
         # tmpl = self.parse_file_header(template, auto_cache=auto_cache, verbose=verbose)
 
-        ### Check for AMOF netCDF file and gather specs ###
-        if template == "auto" and file_path.split('.')[-1] == 'nc':
-            # Look for AMOF Convention string in Conventions global attr, if it exists
-            if ':Conventions' in file_content.cdl:
-                conventions = file_content.cdl.split(':Conventions =')[1].split(';')[0].strip()
-                if "NCAS-AMOF" in conventions or "NCAS-GENERAL" in conventions or "NCAS-AMF" in conventions:
+        ### Check for NCAS data files and gather specs ###
+        # if template and specs are "default" values, check to see if
+        # file is an ncas file (assuming file name starts with instrument name)
+        if (template == "auto" and specs == None and
+            file_path.split("/")[-1].startswith("ncas-")):
+            # find appropriate specs depending on convention
+            if file_path.split(".")[-1] == "nc" and ":Conventions" in file_content.cdl:
+                conventions = file_content.cdl.split(":Conventions =")[1].split(";")[0].strip()
+                # NCAS-GENERAL file
+                if any(name in conventions for name in ["NCAS-GENERAL", "NCAS-AMF", "NCAS-AMOF"]):
                     if verbose:
                         print("\nNCAS-AMOF file detected, finding correct spec files")
                         print("Finding correct AMOF version...")
@@ -232,7 +236,7 @@ def check_file(self, file_path, template="auto", mappings=None, extra_rules=None
                     # check specs exist for that version
                     specs_dir = os.path.join(conf["settings"].get("specs_dir", "./specs"), f"groups/{spec_folder}")
                     if not os.path.exists(specs_dir):
-                        if verbose: print(f"Specs for version {version_number} not found, attempting download...")
+                        if verbose: print(f"Specs for version NCAS-GENERAL-{version_number} not found, attempting download...")
                         try:
                             vocabs_dir = os.path.join(conf["settings"].get("vocabs_dir", "./checksit/vocabs"), f"AMF_CVs/{version_number}")
                             cvs = urllib.request.urlopen(f"https://github.com/ncasuk/AMF_CVs/tree/v{version_number}/AMF_CVs")
@@ -261,7 +265,6 @@ def check_file(self, file_path, template="auto", mappings=None, extra_rules=None
                             sys.exit()
                         except:
                             raise
-                            
 
                     # get deployment mode and data product, to then get specs
                     deployment_mode = file_content.cdl.split(':deployment_mode =')[1].split(';')[0].strip().strip('"')
@@ -272,6 +275,30 @@ def check_file(self, file_path, template="auto", mappings=None, extra_rules=None
                     # don't need to do template check
                     template = "off"
 
+                # NCAS-RADAR (coming soon...)
+                # if "NCAS-Radar" in conventions
+            
+            elif (file_path.split(".")[-1] in ["png", "PNG", "jpg", "JPG", "jpeg", "JPEG"] and
+                  "XMP-photoshop:Instructions" in file_content.global_attrs.keys()):
+                conventions = file_content.global_attrs["XMP-photoshop:Instructions"]
+                if "National Centre for Atmospheric Science Image Metadata Standard" in file_content.global_attrs["XMP-photoshop:Instructions"].replace("\n"," "):
+                    if verbose:
+                        print("\nNCAS-IMAGE file detected, finding correct spec files")
+                        print("Finding correct IMAGE version...")
+                    version_number = conventions.replace("\n"," ").split("Metadata Standard ")[1].split(":")[0]
+                    spec_folder = f"ncas-image-{version_number}"
+                    if verbose: print(f"  {version_number}")
+                    specs_dir = os.path.join(conf["settings"].get("specs_dir", "./specs"), f"groups/{spec_folder}")
+                    if not os.path.exists(specs_dir):
+                        print(f"[ERROR] specs for NCAS-IMAGE {version_number} can not be found.")
+                        print("Aborting...")
+                        sys.exit()
+                    product = file_path.split('/')[-1].split('_')[3]
+                    product_spec = f"{spec_folder}/amof-{product}"
+                    specs = [product_spec, f"{spec_folder}/amof-image-global-attrs"]
+                    template = "off"
+
+
 
         if template == "off":
             tmpl = template
diff --git a/checksit/readers/cdl.py b/checksit/readers/cdl.py
index 5ee68afe..a040e409 100644
--- a/checksit/readers/cdl.py
+++ b/checksit/readers/cdl.py
@@ -78,7 +78,7 @@ def _get_sections(self, lines, split_patterns, start_at):
                 if split_patterns:
                     splitter = split_patterns.popleft()
             else:
-                line_no_comments = re.split(";\s+//.*$", line)[0].strip().rstrip(";").strip()
+                line_no_comments = re.split(r";\s+//.*$", line)[0].strip().rstrip(";").strip()
                 if not line_no_comments.startswith("//"):
                     current.append(line_no_comments)