add parts

.gitignore

@@ -21,7 +21,6 @@ eggs/
 .eggs/
 lib/
 lib64/
-parts/
 sdist/
 var/
 wheels/

buzzword/create.py

@@ -27,7 +27,9 @@
 BUZZWORD_PAGE_SIZE=25
 BUZZWORD_TABLE_SIZE=2000,200
 BUZZWORD_ADD_GOVERNOR=false
-""".format(FULLNAME)
+""".format(
+    FULLNAME
+)
 
 CORPORA = """
 {{
@@ -61,11 +63,11 @@
 for folder in ["csv", "uploads", "example"]:
     os.makedirs(os.path.join(FULLNAME, folder))
 with open(os.path.join(FULLNAME, ".env"), "w") as fo:
-    fo.write(ENV.strip() + '\n')
+    fo.write(ENV.strip() + "\n")
 with open(os.path.join(FULLNAME, "corpora.json"), "w") as fo:
-    fo.write(CORPORA.strip() + '\n')
+    fo.write(CORPORA.strip() + "\n")
 with open(os.path.join(CORPUS_PATH, "001-joke-lion-pun.txt"), "w") as fo:
-    fo.write(CORPUS.strip() + '\n')
+    fo.write(CORPUS.strip() + "\n")
 print("Testing parser: {}->{}-parsed ...".format(CORPUS_PATH, CORPUS_PATH))
 parsed = Corpus(CORPUS_PATH).parse(cons_parser=None)
 

buzzword/parts/about.py

@@ -0,0 +1,13 @@
+import dash_core_components as dcc
+import dash_html_components as html
+from buzzword.parts.nav import navbar
+import os
+import buzzword
+
+
+root = os.path.dirname(os.path.dirname(buzzword.__file__))
+
+with open(os.path.join(root, "docs/about.md"), "r") as fo:
+    text = fo.read()
+
+layout = html.Div([navbar, dcc.Markdown(className="documentation", children=text)])

buzzword/parts/assets/custom.css

@@ -0,0 +1,20 @@
+.loading-main::after{
+  content: "Preparing your workspace...";
+  position: fixed;
+  margin-top: 100px;
+}
+
+/* Hide the 'Loading...' text at startup */
+._dash-loading {
+	display:none;
+}
+
+.no-underline {
+	text-decoration: none;
+	color: blue;
+}
+
+p, ul, ol {
+	padding-bottom: 10px !important;
+	padding-top: 10px !important;
+}

buzzword/parts/building.py

@@ -0,0 +1,12 @@
+import dash_core_components as dcc
+import dash_html_components as html
+from buzzword.parts.nav import navbar
+import os
+import buzzword
+
+root = os.path.dirname(os.path.dirname(buzzword.__file__))
+
+with open(os.path.join(root, "docs/building.md"), "r") as fo:
+    text = fo.read()
+
+layout = html.Div([navbar, dcc.Markdown(className="documentation", children=text)])

buzzword/parts/configure.py

@@ -0,0 +1,168 @@
+# flake8: noqa
+
+"""
+buzz webapp: command-line and .env processing
+"""
+
+import argparse
+import os
+
+from dotenv import load_dotenv
+
+
+def _from_cmdline():
+    """
+    Command line argument parsing
+    """
+    parser = argparse.ArgumentParser(
+        description="Run the buzzword app for a given corpus."
+    )
+
+    parser.add_argument(
+        "-nl",
+        "--load",
+        default=True,
+        action="store_true",
+        required=False,
+        help="Load corpus into memory. Longer startup, faster search.",
+    )
+
+    parser.add_argument(
+        "-r",
+        "--root",
+        default=".",
+        required=False,
+        type=str,
+        nargs="?",
+        help="Space for the tool to store CSVs, uploaded data, etc.",
+    )
+
+    parser.add_argument(
+        "-t", "--title", nargs="?", type=str, required=False, help="Title for app"
+    )
+
+    parser.add_argument(
+        "-d",
+        "--drop-columns",
+        nargs="?",
+        type=str,
+        required=False,
+        help="Dataset columns to remove before loading (comma-separated)",
+    )
+
+    parser.add_argument(
+        "-m",
+        "--max-dataset-rows",
+        nargs="?",
+        type=int,
+        required=False,
+        help="Truncate datasets at this many rows",
+    )
+
+    parser.add_argument(
+        "-s",
+        "--table-size",
+        nargs="?",
+        type=str,
+        required=False,
+        default="2000,200",
+        help="Max table dimensions as str ('nrows,ncolumns')",
+    )
+
+    parser.add_argument(
+        "-p",
+        "--page-size",
+        nargs="?",
+        type=int,
+        default=25,
+        required=False,
+        help="Rows to display per page",
+    )
+
+    parser.add_argument(
+        "-debug",
+        "--debug",
+        default=False,
+        action="store_true",
+        required=False,
+        help="Debug mode",
+    )
+
+    parser.add_argument(
+        "-g",
+        "--add-governor",
+        default=False,
+        action="store_true",
+        required=False,
+        help="Load governor attributes into dataset. Slow to load and uses more memory, but allows more kinds of searching/showing",
+    )
+
+    parser.add_argument(
+        "-e",
+        "--env",
+        nargs="?",
+        type=str,
+        required=False,
+        help="Use .env file to load config, rather than command line",
+    )
+
+    parser.add_argument(
+        "-c",
+        "--corpora-file",
+        default="corpora.json",
+        type=str,
+        nargs="?",
+        help="Path to corpora.json",
+    )
+
+    kwargs = vars(parser.parse_args())
+    if kwargs["drop_columns"] is not None:
+        kwargs["drop_columns"] = kwargs["drop_columns"].split(",")
+    if kwargs["table_size"] is not None:
+        kwargs["table_size"] = [int(i) for i in kwargs["table_size"].split(",")][:2]
+    return kwargs
+
+
+def _configure_buzzword(name):
+    """
+    Configure application. First, look at command line args.
+    If the user wants to use dotenv (--env flag), load from that.
+    If not from main, use dotenv only.
+    """
+    env_path = os.path.join(os.getcwd(), ".env")
+    config = _from_cmdline()
+    if not config["env"]:
+        return config
+    else:
+        env_path = config["env"]
+    return _from_env(env_path)
+
+
+def _from_env(env_path):
+    """
+    Read .env. Should return same as command line, except --env argument
+    """
+    trues = {"1", "true", "True", "Y", "y", "yes", True}
+    load_dotenv(dotenv_path=env_path)
+    drop_columns = os.getenv("BUZZWORD_DROP_COLUMNS")
+    if drop_columns:
+        drop_columns = drop_columns.split(",")
+    table_size = os.getenv("BUZZWORD_TABLE_SIZE")
+    if table_size:
+        table_size = [int(i) for i in table_size.split(",")]
+    max_dataset_rows = os.getenv("BUZZWORD_MAX_DATASET_ROWS")
+    if max_dataset_rows:
+        max_dataset_rows = int(max_dataset_rows)
+
+    return dict(
+        corpora_file=os.getenv("BUZZWORD_CORPORA_FILE", "corpora.json"),
+        root=os.getenv("BUZZWORD_ROOT", "."),
+        debug=os.getenv("BUZZWORD_DEBUG", True) in trues,
+        load=os.getenv("BUZZWORD_LOAD", True) in trues,
+        add_governor=os.getenv("BUZZWORD_ADD_GOVERNOR", False) in trues,
+        title=os.getenv("BUZZWORD_TITLE"),
+        page_size=int(os.getenv("BUZZWORD_PAGE_SIZE", 25)),
+        max_dataset_rows=max_dataset_rows,
+        drop_columns=drop_columns,
+        table_size=table_size,
+    )

buzzword/parts/depgrep.py

@@ -0,0 +1,12 @@
+import dash_core_components as dcc
+import dash_html_components as html
+from buzzword.parts.nav import navbar
+import os
+import buzzword
+
+root = os.path.dirname(os.path.dirname(buzzword.__file__))
+
+with open(os.path.join(root, "docs/depgrep.md"), "r") as fo:
+    text = fo.read()
+
+layout = html.Div([navbar, dcc.Markdown(className="documentation", children=text)])