Merge branch 'snip-import' of github.com:afallou/community into snip-…

…import

.gitignore

@@ -4,6 +4,9 @@ __pycache__
 .idea/
 # Locally generated
 /settings
+# we highly recommended against removing the private folder from the gitignore file
+# the directory is used to store sensitive information such as contacts
+/private
 .vscode/settings.json
 .DS_Store
 *.bak

README.md

@@ -282,7 +282,7 @@ Most lists of words are provided as Talon list files, with an extension of `.tal
 
 Some lists with multiple spoken forms/alternatives are instead provided as CSV files. Some are in the `settings` folder and are not created until you launch Talon with `community` installed.
 
-You can customize common Talon list and CSV files with voice commands: say the word `customize` followed by `abbreviations`, `additional words`, `alphabet`, `homophones`, `search engines`, `Unix utilities`, `websites` or `words to replace`. These open the file in a text editor and move the insertion point to the bottom of the file so you can add to it.
+You can customize common Talon list and CSV files with voice commands: say the word `customize` followed by `abbreviations`, `additional words`, `alphabet`, `homophones`, `search engines`, `Unix utilities`, `websites`, `words to replace`, `contacts json` or `contacts csv`. These open the file in a text editor and move the insertion point to the bottom of the file so you can add to it.
 
 You can also add words to the vocabulary or replacements (words_to_replace) by using the commands in [edit_vocabulary.talon](core/vocabulary/edit_vocabulary.talon).
 

core/README.md

@@ -4,6 +4,7 @@ This folder contains `edit_settings.talon`, which has a command to open various
 
 - `abbreviate` has a command for the use of abbreviations
 - `app_switcher` does not have commands but has the implementation of functions that allow for switching between applications
+- `contacts` has captures for inserting contact information including name and email.
 - `edit` has commands for navigating and editing text with copy, paste, etc., as well as commands for zooming in and out
 - `file_extension` has a command for simpler spoken forms of file and website name extensions
 - `help` has commands to open various help menus, as described in the top level [README](https://github.com/talonhub/community?tab=readme-ov-file#getting-started-with-talon)

core/contacts/README.md

@@ -0,0 +1,61 @@
+# Contacts
+
+This directory provides a versatile `<user.prose_contact>` capture that can be
+used to insert names and email addresses using a suffix. The contact list may be
+provided in the private directory via the `contacts.json` file, the `contacts.csv` file, or both.
+
+Here is an example contacts.json:
+
+```json
+[
+  {
+    "email": "john.doe@example.com",
+    "full_name": "Jonathan Doh: Jonathan Doe",
+    "nicknames": ["Jon", "Jah Nee: Jonny"]
+  }
+]
+```
+
+Note that for either full_name or nicknames, pronunciation can be provided via
+the standard Talon list format of "[pronunciation]: [name]". Pronunciation for
+the first name is automatically extracted from pronunciation for the full name,
+if there are the same number of name parts in each. Pronunciation can be
+overridden for the first name by adding a nickname with matching written form.
+
+To refer to this contact, you could say:
+
+- Jonathan Doh email -> john.doe@example.com
+- Jonathan email -> john.doe@example.com
+- Jah Nee email -> john.doe@example.com
+- Jah Nee name -> Jonny
+- Jonathan Doh name -> Jonathan Doe
+- Jon last name -> Doe
+- Jon full name -> Jonathan Doe
+- Jon names -> Jon's
+- Jon full names -> Jonathan Doe's
+
+The CSV format provides only email and full name functionality:
+
+```csv
+Name,Email
+John Doe,jon.doe@example.com
+Jane Doe,jane.doe@example.com
+```
+
+The advantage of the CSV format is that it is easily exported. If both the CSV
+and JSON are present, they will be merged based on email addresses. This makes
+it easy to use an exported CSV and maintain nicknames in the JSON. For example,
+to export from Gmail, go to https://contacts.google.com/, then click "Frequently
+contacted", then "Export". Then run:
+
+```bash
+cat contacts.csv | python -c "import csv; import sys; w=csv.writer(sys.stdout); [w.writerow([row['First Name'] + ' ' + row['Last Name'], row['E-mail 1 - Value']]) for row in csv.DictReader(sys.stdin)]"
+```
+
+In case of name conflicts (e.g. two people named John), the first instance will
+be preferred, with all JSON contacts taking precedence over CSV. If you wish to
+refer to both, use the pronunciation to differentiate, using a nickname to
+override the first name pronunciation if desired. For example, you might add
+"John S: John" and "John D: John" as nicknames to the two different Johns. This
+is also an effective way to handle name homophones such as John and Jon, which
+would otherwise be resolved arbitrarily by the speech engine.

core/contacts/contacts.py

@@ -0,0 +1,293 @@
+import json
+import logging
+from dataclasses import dataclass
+
+from talon import Context, Module
+
+from ..user_settings import track_csv_list, track_file
+
+mod = Module()
+ctx = Context()
+
+mod.list("contact_names", desc="Contact first names, full names, and nicknames.")
+mod.list("contact_emails", desc="Maps names to email addresses.")
+mod.list("contact_full_names", desc="Maps names to full names.")
+
+
+@dataclass
+class Contact:
+    email: str
+    full_name: str
+    nicknames: list[str]
+    pronunciations: dict[str, str]
+
+    @classmethod
+    def from_json(cls, contact):
+        email = contact.get("email")
+        if not email:
+            logging.error(f"Skipping contact missing email: {contact}")
+            return None
+
+        # Handle full name with potential pronunciation
+        full_name_raw = contact.get("full_name", "")
+        pronunciations = {}
+        if ":" in full_name_raw:
+            pronunciation, full_name = [x.strip() for x in full_name_raw.split(":", 1)]
+            if (
+                full_name in pronunciations
+                and pronunciations[full_name] != pronunciation
+            ):
+                logging.info(
+                    f"Multiple pronunciations found for '{full_name}'; using '{pronunciation}'"
+                )
+            pronunciations[full_name] = pronunciation
+
+            # Add pronunciation for each component of the name.
+            pron_parts = pronunciation.split()
+            name_parts = full_name.split()
+            if len(pron_parts) == len(name_parts):
+                for pron, name in zip(pron_parts, name_parts):
+                    if name in pronunciations and pronunciations[name] != pron:
+                        logging.info(
+                            f"Multiple different pronunciations found for '{name}' in "
+                            f"{full_name_raw}; using '{pron}'"
+                        )
+                    pronunciations[name] = pron
+            else:
+                logging.info(
+                    f"Pronunciation parts don't match name parts for '{full_name_raw}; skipping them.'"
+                )
+        else:
+            full_name = full_name_raw
+
+        # Handle nicknames with potential pronunciations
+        nicknames = []
+        for nickname_raw in contact.get("nicknames", []):
+            if ":" in nickname_raw:
+                pronunciation, nickname = [
+                    x.strip() for x in nickname_raw.split(":", 1)
+                ]
+                if (
+                    nickname in pronunciations
+                    and pronunciations[nickname] != pronunciation
+                ):
+                    logging.info(
+                        f"Multiple different pronunciations found for '{nickname}' in "
+                        f"contact {email}; using '{pronunciation}'"
+                    )
+                pronunciations[nickname] = pronunciation
+                nicknames.append(nickname)
+            else:
+                nicknames.append(nickname_raw)
+
+        return Contact(
+            email=email,
+            full_name=full_name,
+            nicknames=nicknames,
+            pronunciations=pronunciations,
+        )
+
+
+csv_contacts: list[Contact] = []
+json_contacts: list[Contact] = []
+
+
+@track_csv_list("contacts.csv", headers=("Name", "Email"), default={}, private=True)
+def on_contacts_csv(values):
+    global csv_contacts
+    csv_contacts = []
+    for email, full_name in values.items():
+        if not email:
+            logging.error(f"Skipping contact missing email: {full_name}")
+            continue
+        csv_contacts.append(
+            Contact(email=email, full_name=full_name, nicknames=[], pronunciations={})
+        )
+    reload_contacts()
+
+
+@track_file("contacts.json", default="[]", private=True)
+def on_contacts_json(f):
+    global json_contacts
+    try:
+        contacts = json.load(f)
+    except Exception:
+        logging.exception("Error parsing contacts.json")
+        return
+
+    json_contacts = []
+    for contact in contacts:
+        try:
+            parsed_contact = Contact.from_json(contact)
+            if parsed_contact:
+                json_contacts.append(parsed_contact)
+        except Exception:
+            logging.exception(f"Error parsing contact: {contact}")
+    reload_contacts()
+
+
+def create_pronunciation_to_name_map(contact):
+    result = {}
+    if contact.full_name:
+        result[contact.pronunciations.get(contact.full_name, contact.full_name)] = (
+            contact.full_name
+        )
+        # Add pronunciation mapping for first name only
+        first_name = contact.full_name.split()[0]
+        result[contact.pronunciations.get(first_name, first_name)] = first_name
+    for nickname in contact.nicknames:
+        result[contact.pronunciations.get(nickname, nickname)] = nickname
+    return result
+
+
+def reload_contacts():
+    csv_by_email = {contact.email: contact for contact in csv_contacts}
+    json_by_email = {contact.email: contact for contact in json_contacts}
+    # Merge the CSV and JSON contacts. Maintain order of contacts with JSON first.
+    merged_contacts = []
+    for email in json_by_email | csv_by_email:
+        csv_contact = csv_by_email.get(email)
+        json_contact = json_by_email.get(email)
+
+        if csv_contact and json_contact:
+            # Prefer JSON data but use CSV name if JSON name is empty
+            full_name = json_contact.full_name or csv_contact.full_name
+            merged_contacts.append(
+                Contact(
+                    email=email,
+                    full_name=full_name,
+                    nicknames=json_contact.nicknames,
+                    pronunciations=json_contact.pronunciations,
+                )
+            )
+        else:
+            # Use whichever contact exists
+            merged_contacts.append(json_contact or csv_contact)
+
+    contact_names = {}
+    contact_emails = {}
+    contact_full_names = {}
+    # Iterate in reverse so that the first contact with a name is used.
+    for contact in reversed(merged_contacts):
+        pronunciation_map = create_pronunciation_to_name_map(contact)
+        for pronunciation, name in pronunciation_map.items():
+            contact_names[pronunciation] = name
+            contact_emails[pronunciation] = contact.email
+            if contact.full_name:
+                contact_full_names[pronunciation] = contact.full_name
+
+    ctx.lists["user.contact_names"] = contact_names
+    ctx.lists["user.contact_emails"] = contact_emails
+    ctx.lists["user.contact_full_names"] = contact_full_names
+
+
+def first_name_from_full_name(full_name: str):
+    return full_name.split(" ")[0]
+
+
+def last_name_from_full_name(full_name: str):
+    return full_name.split(" ")[-1]
+
+
+def username_from_email(email: str):
+    return email.split("@")[0]
+
+
+def make_name_possessive(name: str):
+    return f"{name}'s"
+
+
+@mod.capture(
+    rule="{user.contact_names} name",
+)
+def prose_name(m) -> str:
+    return m.contact_names
+
+
+@mod.capture(
+    rule="{user.contact_names} names",
+)
+def prose_name_possessive(m) -> str:
+    return make_name_possessive(m.contact_names)
+
+
+@mod.capture(
+    rule="{user.contact_emails} email [address]",
+)
+def prose_email(m) -> str:
+    return m.contact_emails
+
+
+@mod.capture(
+    rule="{user.contact_emails} (username | L dap)",
+)
+def prose_username(m) -> str:
+    return username_from_email(m.contact_emails)
+
+
+@mod.capture(
+    rule="{user.contact_full_names} full name",
+)
+def prose_full_name(m) -> str:
+    return m.contact_full_names
+
+
+@mod.capture(
+    rule="{user.contact_full_names} full names",
+)
+def prose_full_name_possessive(m) -> str:
+    return make_name_possessive(m.contact_full_names)
+
+
+@mod.capture(
+    rule="{user.contact_full_names} first name",
+)
+def prose_first_name(m) -> str:
+    return first_name_from_full_name(m.contact_full_names)
+
+
+@mod.capture(
+    rule="{user.contact_full_names} first names",
+)
+def prose_first_name_possessive(m) -> str:
+    return make_name_possessive(first_name_from_full_name(m.contact_full_names))
+
+
+@mod.capture(
+    rule="{user.contact_full_names} last name",
+)
+def prose_last_name(m) -> str:
+    return last_name_from_full_name(m.contact_full_names)
+
+
+@mod.capture(
+    rule="{user.contact_full_names} last names",
+)
+def prose_last_name_possessive(m) -> str:
+    return make_name_possessive(last_name_from_full_name(m.contact_full_names))
+
+
+@mod.capture(
+    rule="(hi | high) {user.contact_names} [name]",
+)
+def prose_contact_snippet(m) -> str:
+    return f"hi {m.contact_names}"
+
+
+@mod.capture(
+    rule=(
+        "<user.prose_name> "
+        "| <user.prose_name_possessive> "
+        "| <user.prose_email> "
+        "| <user.prose_username> "
+        "| <user.prose_full_name> "
+        "| <user.prose_full_name_possessive> "
+        "| <user.prose_first_name> "
+        "| <user.prose_first_name_possessive> "
+        "| <user.prose_last_name>"
+        "| <user.prose_last_name_possessive>"
+        "| <user.prose_contact_snippet>"
+    ),
+)
+def prose_contact(m) -> str:
+    return m[0]

core/edit_text_file/edit_text_file_list.talon-list

@@ -13,3 +13,5 @@ unix utilities: tags/terminal/unix_utility.talon-list
 abbreviations: settings/abbreviations.csv
 file extensions: settings/file_extensions.csv
 words to replace: settings/words_to_replace.csv
+contacts json: settings/contacts.json
+contacts csv: settings/contacts.csv