Merge cloud branch

Author: jalakoo

Pipfile

@@ -15,6 +15,8 @@ streamlit-player = "*"
 requests = "*"
 pandas = "*"
 random-address = "*"
+openai = "*"
+numpy = "*"
 
 [dev-packages]
 mock-generators = {editable = true, path = "."}

Pipfile.lock

@@ -1,9 +1,12 @@
 import streamlit as st
 from constants import *
+from tabs.ideate_tab import ideate_tab
 from tabs.importing_tab import import_tab
 from tabs.design_tab import design_tab
 from tabs.data_importer import data_importer_tab
 from tabs.tutorial import tutorial_tab
+from tabs.getting_help import get_help_tab
+
 from config import setup_logging, preload_state, load_generators_to_streamlit
 
 # SETUP
@@ -22,18 +25,24 @@
 
 # Streamlit runs from top-to-bottom from tabs 1 through 8. This is essentially one giant single page app.  Earlier attempt to use Streamlit's multi-page app functionality resulted in an inconsistent state between pages.
 
-t0, t1, t2, t5 = st.tabs([
-    "⓪ Tutorial",
-    "① Design",
-    "② Generate",
-    "③ Data Importer"
+t0, t1, t2, t3, t4, t5 = st.tabs([
+    "⓪ Getting Started",
+    "① Ideate",
+    "② Design",
+    "③ Generate",
+    "④ Data Importer",
+    "Ⓘ Info"
 ])
 
 with t0:
     tutorial_tab()
 with t1:
-    design_tab()
+    ideate_tab()
 with t2:
+    design_tab()
+with t3:
     import_tab()
+with t4:
+    data_importer_tab()
 with t5:
-    data_importer_tab()
+    get_help_tab()

mock_generators/app.py

@@ -1,6 +1,15 @@
-from random_address import real_random_address_by_state
+from random_address import real_random_address
 import random
 
+def generate(args: list[any]):
+    # Generate a dictionary with valid random address information
+    # states = [
+    #     "AL", "AR", "CA", "CO", "CT", "DC", "FL", "GA", "HI", "KY", "MA" "MD", "TN", "TX", "OK", "VT"
+    # ]
+    # state_code = random.choice(states)
+    # return real_random_address_by_state(state_code)
+    return real_random_address()
+
 def generate(args: list[any]):
     # Generate a dictionary with valid random address information
     states = [

mock_generators/generators/address_usa.py

@@ -0,0 +1,201 @@
+# For converting data from various forms to other forms
+from streamlit_agraph import agraph, Node, Edge, Config
+import logging
+import json
+
+def random_coordinates_for(
+        number: int):
+    import numpy as np
+
+    # Generate random coordinates
+    min_range = 0
+    max_width = number * 300
+    x_values = np.random.uniform(min_range, max_width, number)
+    y_values = np.random.uniform(min_range, max_width, number)
+
+    # Create list of (x, y) coordinates
+    coordinates = list(zip(x_values, y_values))
+    return coordinates
+
+
+def agraph_data_from_response(response: str)->tuple[any, any, any]:
+    # Returns agraph compatible nodes, edges, and config
+    logging.debug(f'Response: {response}')
+    # Response will be a list of 3 item tuples
+    try:
+        answers = json.loads(response)
+    except Exception as e:
+        logging.error(e)
+        return None, None, None
+    if isinstance(answers, list) == False:
+        logging.error(f'Response could not be converted to list, got {type(answers)} instead.')
+        return None, None, None
+    nodes = []
+    # node_labels = []
+    edges = []
+    for idx, item in enumerate(answers):
+        if item is None or len(item) != 3:
+            continue
+        n1_label = item[0]
+        r = item[1]
+        n2_label = item[2]
+
+        # We only want to add agraph nodes with the same label once in our return
+
+        # Gross but works
+        add_n1 = True
+        for node in nodes:
+            if node.label == n1_label:
+                add_n1 = False
+        if add_n1:
+            nodes.append(Node(id=n1_label, label=n1_label))
+
+        add_n2 = True
+        for node in nodes:
+            if node.label == n2_label:
+                add_n2 = False
+        if add_n2:
+            nodes.append(Node(id=n2_label, label=n2_label))
+
+        # agraph requires source and target ids to use what we consider labels
+        edge = Edge(source=n1_label, target=n2_label, label=r)
+        edges.append(edge)
+    config = Config(
+        width=800, 
+        height=800, 
+        backgroundColor="#000000",
+        directed=True)
+    return (nodes, edges, config)
+
+
+def convert_agraph_nodes_to_arrows_nodes(
+        agraph_nodes: list
+)-> list[dict]:
+    # Convert agraph nodes to arrows nodes
+    arrows_nodes = []
+
+    # Generate random coordinates to init new arrows nodes with - since we can't extract the location data from agraph
+    coordinates = random_coordinates_for(len(agraph_nodes))
+
+    for nidx, node in enumerate(agraph_nodes):
+        new_node = convert_agraph_node_to_arrows_node(
+            nidx, node, coordinates[nidx][0], coordinates[nidx][1])
+        arrows_nodes.append(new_node)
+    return arrows_nodes
+
+def convert_agraph_node_to_arrows_node(
+        idx, 
+        node,
+        x,
+        y):
+    # Convert agraph node to arrows node
+    arrows_node = {
+        "id": f'n{idx+1}',
+        "caption": node.label,
+        "position": {
+            "x": x,
+            "y": y,
+        },
+        "labels":[],
+        "style": {},
+        "properties": {}
+    }
+    return arrows_node
+
+def convert_agraph_edge_to_arrows_relationship(
+        idx, 
+        edge, 
+        arrows_nodes: list):
+    # Example: {'source': 'People', 'from': 'People', 'to': 'Cars', 'color': '#F7A7A6', 'label': 'DRIVE'}
+    source_node_label = edge.source
+    target_node_label = edge.to
+    source_node_id = None
+    target_node_id = None
+
+    for node in arrows_nodes:
+        if node['caption'] == source_node_label:
+            source_node_id = node['id']
+        if node['caption'] == target_node_label:
+            target_node_id = node['id']
+
+    if source_node_id is None or target_node_id is None:
+        node_info = [node.__dict__ for node in arrows_nodes]
+        logging.error(f'Could not find source or target node for edge {edge.__dict__} from nodes: {node_info}')
+        return None
+    edge_type = edge.label
+    arrows_relationship = {
+        "id": f'n{idx+1}',
+        "fromId": source_node_id,
+        "toId": target_node_id,
+        "type": edge_type,
+        "properties": {},
+        "style": {}
+    }
+    return arrows_relationship
+
+def convert_agraph_to_arrows(agraph_nodes, agraph_edges):
+    arrows_nodes = convert_agraph_nodes_to_arrows_nodes(agraph_nodes)
+
+    arrows_relationships = []
+    for eidx, edge in enumerate(agraph_edges):
+        new_relationship = convert_agraph_edge_to_arrows_relationship(eidx, edge, arrows_nodes=arrows_nodes)
+        arrows_relationships.append(new_relationship)
+    arrows_json = {
+        "nodes": arrows_nodes,
+        "relationships": arrows_relationships,
+          "style": {
+            "font-family": "sans-serif",
+            "background-color": "#ffffff",
+            "background-image": "",
+            "background-size": "100%",
+            "node-color": "#ffffff",
+            "border-width": 4,
+            "border-color": "#000000",
+            "radius": 50,
+            "node-padding": 5,
+            "node-margin": 2,
+            "outside-position": "auto",
+            "node-icon-image": "",
+            "node-background-image": "",
+            "icon-position": "inside",
+            "icon-size": 64,
+            "caption-position": "inside",
+            "caption-max-width": 200,
+            "caption-color": "#000000",
+            "caption-font-size": 50,
+            "caption-font-weight": "normal",
+            "label-position": "inside",
+            "label-display": "pill",
+            "label-color": "#000000",
+            "label-background-color": "#ffffff",
+            "label-border-color": "#000000",
+            "label-border-width": 4,
+            "label-font-size": 40,
+            "label-padding": 5,
+            "label-margin": 4,
+            "directionality": "directed",
+            "detail-position": "inline",
+            "detail-orientation": "parallel",
+            "arrow-width": 5,
+            "arrow-color": "#000000",
+            "margin-start": 5,
+            "margin-end": 5,
+            "margin-peer": 20,
+            "attachment-start": "normal",
+            "attachment-end": "normal",
+            "relationship-icon-image": "",
+            "type-color": "#000000",
+            "type-background-color": "#ffffff",
+            "type-border-color": "#000000",
+            "type-border-width": 0,
+            "type-font-size": 16,
+            "type-padding": 5,
+            "property-position": "outside",
+            "property-alignment": "colon",
+            "property-color": "#000000",
+            "property-font-size": 16,
+            "property-font-weight": "normal"
+        }
+    }
+    return arrows_json
+    

mock_generators/logic/agraph_conversions.py

@@ -2,6 +2,7 @@
 from models.generator import Generator, GeneratorType
 import logging
 import json
+import re
 
 # ORIGINAL GENERATOR ASSIGNMENT
 def actual_generator_for_raw_property(
@@ -94,11 +95,39 @@ def is_int(value: str) -> bool:
 
 def is_float(value: str) -> bool:
     try:
-        f = float(value)
+        _ = float(value)
         return True
     except ValueError:
         return False
 
+def find_longest_float_precision(float_list):
+    max_precision = 0
+    for num in float_list:
+        num_str = str(num)
+        if '.' in num_str:
+            decimal_part = num_str.split('.')[1]
+            precision = len(decimal_part)
+            max_precision = max(max_precision, precision)
+    return max_precision
+
+def extract_numbers(string):
+    # Use regex to find all number patterns in the string
+    number_list = re.findall(r"-?\d+(?:\.\d+)?", string)
+
+    # Convert the extracted strings to appropriate number types
+    number_list = [int(num) if num.isdigit() else float(num) for num in number_list]
+
+    return number_list
+
+def numbers_list_from(string):
+    # numbers = []
+    # ranges = string.split('-')
+    # for r in ranges:
+    #     numbers.extend(extract_numbers(r))
+    # return numbers 
+    options = re.split(r'(?<=[^-])-', string)
+    return options
+
 def literal_generator_from_value(
         value: str,
         generators: list[Generator]
@@ -134,9 +163,10 @@ def literal_generator_from_value(
     # Original specificaion took stringified JSON objects to notate generator and args to use. We're going to convert matching literal values to appropriate generators
 
     # Default is to use the literal generator
-    result = {
-        "string": [value]
-    }
+    result = None
+    # result = {
+    #     "string": [value]
+    # }
 
     # Check if value is an int or float
     if is_int(value):
@@ -145,32 +175,31 @@ def literal_generator_from_value(
             "int": [integer]
         }
 
-    if is_float(value):
+    if result is None and is_float(value):
         f = float(value)
         result = {
             "float": [f]
         }
 
     # NOTE: Not currently handling complex literals
 
-    # Check if value is a range of ints or floats
-    r = value.split("-")
-    if len(r) == 2:
-        # Single dash in string, possibly a literal range
-        values = [r[0], r[1]]
-        if all_ints(values):
-            result = {
-                "int_range": [int(r[0]), int(r[1])]
-            }
-        elif some_floats(values):
-            # Float range function expects 3 args - this one seems more sensible than other functions
-            result = {
-                "float_range": [float(r[0]), float(r[1]), 2]
-            }
- 
+    # Check if value is a range of positive ints or floats
+    if result is None:
+        numbers = numbers_list_from(value) 
+        if len(numbers) == 2:
+        # Check for correctly formatted int or float range string
+            precision = find_longest_float_precision(numbers)
+            if precision == 0:
+                    result = {
+                        "int_range": [int(numbers[0]), int(numbers[1])]
+                    }
+            else:
+                    result = {
+                        "float_range": [float(numbers[0]), float(numbers[1]), precision]
+                    }
 
     # Check for literal list of ints, floats, or strings
-    if value.startswith('[') and value.endswith(']'):
+    if result is None and value.startswith('[') and value.endswith(']'):
         values = value[1:-1].split(',')
         # Generators take a strange format where the args are always a string - including # lists of other data, like ints, floats. ie ["1,2,3"] is an expected arg type
         # because certain generators could take multiple args from different text fields
@@ -193,6 +222,10 @@ def literal_generator_from_value(
                 "string_from_list": values
             }
 
+    if result is None:
+        result = {
+            "string": [value]
+        }
     # Package and return from legacy process
     actual_string = json.dumps(result)
     return actual_generator_for_raw_property(actual_string, generators)