Merge branch 'main' into dthomas_meilisearch

Dockerfile

@@ -0,0 +1,11 @@
+FROM python:latest
+
+RUN apt-get update && apt-get install -y libgl1-mesa-glx ghostscript && apt-get clean && rm -rf /var/lib/apt/lists
+COPY . .
+#COPY config-server.yml config.yml
+RUN pip3 install -r requirements.txt
+
+CMD ["python3", "run.py"]
+EXPOSE 5000
+EXPOSE 8000
+EXPOSE 9000

compose.yml

@@ -0,0 +1,13 @@
+services:
+  meilisearch:
+    image: "getmeili/meilisearch:v1.6.2"
+    ports:
+      - "7700:7700"
+    environment:
+      MEILI_MASTER_KEY: fluffybunnyrabbit
+      MEILI_NO_ANALYTICS: true
+    volumes:
+      - "meili_data:/meili_data"
+
+volumes:
+  meili_data:

get_specs.py

@@ -157,7 +157,7 @@ def get_multi(partnums):
             bar(skipped=True)
             fprint("Parsing Datasheet contents of " + partnum)
             bar.text = "Parsing Datasheet contents of " + partnum + ".pdf..."
-            read_datasheet.parse(path, output_dir)
+            read_datasheet.parse(path, output_dir, partnum)
             bar(skipped=False)
 
         def __downloaded_datasheet(partnum, path, output_dir):
@@ -166,7 +166,7 @@ def get_multi(partnums):
             bar(skipped=False)
             fprint("Parsing Datasheet contents of " + partnum)
             bar.text = "Parsing Datasheet contents of " + partnum + ".pdf..."
-            read_datasheet.parse(path, output_dir)
+            read_datasheet.parse(path, output_dir, partnum)
             bar(skipped=False)
 
         for partnum in partnums:
@@ -227,7 +227,7 @@ def get_multi(partnums):
 
 
 if __name__ == "__main__":
-    partnums = ["10GXS12", "RST 5L-RKT 5L-949", 
+    partnums = ["7958A", "10GXS12", "RST 5L-RKT 5L-949", 
 "10GXS13",
 "10GXW12",
 "10GXW13",

read_datasheet.py

@@ -9,8 +9,10 @@ from PIL import Image
 import io
 import json
 from util import fprint
+import uuid
+from util import run_cmd
 
-def parse(filename, output_dir):
+def parse(filename, output_dir, partnum):
 
     # Extract table data
 
@@ -163,18 +165,77 @@ def parse(filename, output_dir):
 
         previous_table = table_name
     
+    # remove multi-line values that occasionally squeak through
+    def replace_newlines_in_dict(d):
+        for key, value in d.items():
+            if isinstance(value, str):
+                # Replace \n with " " if the value is a string
+                d[key] = value.replace('\n', ' ')
+            elif isinstance(value, dict):
+                # Recursively call the function if the value is another dictionary
+                replace_newlines_in_dict(value)
+        return d
     
-    fprint(tables)
+    tables = replace_newlines_in_dict(tables)
-    with open(output_dir + "/tables.json", 'w') as json_file:
+
-        json.dump(tables, json_file)
+    # summary
+
+    output_table = dict()
+    output_table["partnum"] = partnum
+    id = str(uuid.uuid4())
+    output_table["id"] = id
+    #output_table["position"] = id
+    #output_table["brand"] = brand
+    output_table["fullspecs"] = tables
+    output_table["searchspecs"] = {"partnum": partnum, **flatten(tables)}
+    
+    output_table["searchspecs"]["id"] = id
     
 
 
+    print(output_table)
+
+    run_cmd("rm " + output_dir + "/*.json") # not reliable!
+    with open(output_dir + "/" + output_table["searchspecs"]["id"] + ".json", 'w') as json_file:
+        json.dump(output_table["searchspecs"], json_file)
+
+    return output_table
 
 
-    return tables
+def flatten(tables):
+    def convert_to_number(s):
+        try:
+            # First, try converting to an integer.
+            return int(s)
+        except ValueError:
+            # If that fails, try converting to a float.
+            try:
+                return float(s)
+            except ValueError:
+                # If it fails again, return the original string.
+                return s
+    out = dict()
+    print("{")
+    for table in tables.keys():
+        for key in tables[table].keys():
+            if len(key) < 64:
+                keyname = key
+            else:
+                keyname = key[0:64]
+
+            fullkeyname = (table + ": " + keyname).replace(".","")
+            if type(tables[table][key]) is not tuple:
+                out[fullkeyname] = convert_to_number(tables[table][key])
+                print("\"" + keyname + "\":", "\"" + str(out[fullkeyname]) + "\",")
+            elif len(tables[table][key]) == 1:
+                out[fullkeyname] = convert_to_number(tables[table][key][0])
+                
+                print("\"" + keyname + "\":", "\"" + str(out[fullkeyname]) + "\",")
+
+    print("}")
+    return out
 
     
 
 if __name__ == "__main__":
-    parse("test2.pdf", "10GXS13")
+    parse("test2.pdf", "cables/10GXS13", "10GXS13")

requirements.txt

@@ -5,7 +5,7 @@ pypdf2==2.12.1
 alive-progress
 requests
 git+https://github.com/Byeongdulee/python-urx.git
-psycopg2
+meilisearch
 pyyaml
 Flask
 selenium

util.py

@@ -123,7 +123,7 @@ class Logger(object):
         self.terminal = sys.stdout
 
     def write(self, message):
-        self.log.write(message)
+        #self.log.write(message)
         #close(filename)
         #self.log = open(filename, "a")
         try: