Compare commits

...

3 Commits

Author SHA1 Message Date
Hocuri
2580ec2201 --wip-- [skip ci] 2024-04-24 16:45:44 +02:00
Hocuri
d2014f6038 Make parsing of name, addr, and timestamp work 2024-04-19 23:07:12 +02:00
Hocuri
ba6f3291ea Add basic parsing code 2024-04-16 19:11:15 +02:00
4 changed files with 199 additions and 12 deletions

13
Cargo.lock generated
View File

@@ -1227,9 +1227,11 @@ dependencies = [
[[package]]
name = "deltachat-contact-tools"
version = "0.1.0"
version = "0.0.0"
dependencies = [
"anyhow",
"chrono",
"ical",
"once_cell",
"regex",
"rusqlite",
@@ -2562,6 +2564,15 @@ dependencies = [
"cc",
]
[[package]]
name = "ical"
version = "0.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9b7cab7543a8b7729a19e2c04309f902861293dcdae6558dfbeb634454d279f6"
dependencies = [
"thiserror",
]
[[package]]
name = "idea"
version = "0.5.1"

View File

@@ -47,7 +47,7 @@ async_zip = { version = "0.0.12", default-features = false, features = ["deflate
backtrace = "0.3"
base64 = "0.22"
brotli = { version = "5", default-features=false, features = ["std"] }
chrono = { version = "0.4.37", default-features=false, features = ["clock", "std"] }
chrono = { workspace = true }
email = { git = "https://github.com/deltachat/rust-email", branch = "master" }
encoded-words = { git = "https://github.com/async-email/encoded-words", branch = "master" }
escaper = "0.1"
@@ -168,7 +168,8 @@ harness = false
anyhow = "1"
once_cell = "1.18.0"
regex = "1.10"
rusqlite = { version = "0.31" }
rusqlite = "0.31"
chrono = { version = "0.4.37", default-features=false, features = ["clock", "std"] }
[features]
default = ["vendored"]

View File

@@ -1,8 +1,8 @@
[package]
name = "deltachat-contact-tools"
version = "0.1.0"
version = "0.0.0" # No semver-stable versioning
edition = "2021"
description = "Contact-related tools, like parsing vcards and sanitizing name and address"
description = "Contact-related tools, like parsing vcards and sanitizing name and address. Meant for internal use in the deltachat crate."
license = "MPL-2.0"
# TODO maybe it should be called "deltachat-text-utils" or similar?
@@ -13,6 +13,8 @@ anyhow = { workspace = true }
once_cell = { workspace = true }
regex = { workspace = true }
rusqlite = { workspace = true } # Needed in order to `impl rusqlite::types::ToSql for EmailAddress`. Could easily be put behind a feature.
chrono = { workspace = true }
ical = { version = "0.11.0", default-features = false, features = ["vcard"] }
[dev-dependencies]
anyhow = { workspace = true, features = ["backtrace"] } # Enable `backtrace` feature in tests.

View File

@@ -29,10 +29,81 @@ use std::fmt;
use std::ops::Deref;
use anyhow::bail;
use anyhow::Context as _;
use anyhow::Result;
use chrono::DateTime;
use once_cell::sync::Lazy;
use regex::Regex;
// TODOs to clean up:
// - Check if sanitizing is done correctly everywhere
// - Apply lints everywhere (https://doc.rust-lang.org/cargo/reference/workspaces.html#the-lints-table)
#[derive(Debug, Default)]
pub struct Contact {
addr: String,
display_name: String,
key: String,
profile_photo: String,
timestamp: u64,
}
pub fn parse_vcard(vcard: String) -> Result<Vec<Contact>> {
let reader = ical::VcardParser::new(vcard.as_bytes());
let mut contacts = Vec::new();
for vcard_contact in reader {
let vcard_contact = vcard_contact?; // TODO should just continue with the next contact
let mut new_contact = Contact::default();
let mut display_name = None;
let mut addr = None;
let mut key = None;
let mut photo = None;
let mut timestamp = None;
for property in vcard_contact.properties {
match &*property.name.to_lowercase() {
"email" => addr = addr.or(property.value),
"fn" => display_name = display_name.or(property.value),
"key" => key = key.or(dbg!(property).value), // TODO hmmm, the ical crate can apparently only parse version 3.0
"photo" => photo = photo.or(property.value),
"rev" => {
timestamp = timestamp.or(property.value);
}
_ => {}
}
}
let (display_name, addr) =
sanitize_name_and_addr(&display_name.unwrap_or_default(), &addr.unwrap_or_default());
new_contact.display_name = display_name;
new_contact.addr = addr;
if let Some(key) = key {
if let Some(base64_key) = key
.strip_prefix("PGP;ENCODING=BASE64:")
.or(key.strip_prefix("TYPE=PGP;ENCODING=b:"))
.or(key.strip_prefix("data:application/pgp-keys;base64,"))
{
new_contact.key = base64_key.to_string();
}
}
if let Some(photo) = photo {
if let Some(base64_photo) = photo
.strip_prefix("PGP;ENCODING=BASE64:")
.or(photo.strip_prefix("TYPE=PGP;ENCODING=b:"))
.or(photo.strip_prefix("data:application/pgp-keys;base64,"))
{}
}
contacts.push(new_contact);
}
Ok(contacts)
}
/// Valid contact address.
#[derive(Debug, Clone)]
pub struct ContactAddress(String);
@@ -81,14 +152,10 @@ impl rusqlite::types::ToSql for ContactAddress {
/// Make the name and address
pub fn sanitize_name_and_addr(name: &str, addr: &str) -> (String, String) {
static ADDR_WITH_NAME_REGEX: Lazy<Regex> = Lazy::new(|| Regex::new("(.*)<(.*)>").unwrap());
if let Some(captures) = ADDR_WITH_NAME_REGEX.captures(addr.as_ref()) {
let (name, addr) = if let Some(captures) = ADDR_WITH_NAME_REGEX.captures(addr.as_ref()) {
(
if name.is_empty() {
strip_rtlo_characters(
&captures
.get(1)
.map_or("".to_string(), |m| normalize_name(m.as_str())),
)
strip_rtlo_characters(captures.get(1).map_or("", |m| m.as_str()))
} else {
strip_rtlo_characters(name)
},
@@ -97,8 +164,21 @@ pub fn sanitize_name_and_addr(name: &str, addr: &str) -> (String, String) {
.map_or("".to_string(), |m| m.as_str().to_string()),
)
} else {
(strip_rtlo_characters(name), addr.to_string())
(
strip_rtlo_characters(&normalize_name(name)),
addr.to_string(),
)
};
let mut name = normalize_name(&name);
// If the 'display name' is just the address, remove it:
// Otherwise, the contact would sometimes be shown as "alice@example.com (alice@example.com)" (see `get_name_n_addr()`).
// If the display name is empty, DC will just show the address when it needs a display name.
if name == addr {
name = "".to_string();
}
(name, addr)
}
/// Normalize a name.
@@ -232,6 +312,65 @@ impl rusqlite::types::ToSql for EmailAddress {
mod tests {
use super::*;
#[test]
fn test_thunderbird() {
let contacts = parse_vcard(
"BEGIN:VCARD
VERSION:4.0
FN:'Alice Mueller'
EMAIL;PREF=1:alice.mueller@posteo.de
UID:a8083264-ca47-4be7-98a8-8ec3db1447ca
END:VCARD
BEGIN:VCARD
VERSION:4.0
FN:'bobzzz@freenet.de'
EMAIL;PREF=1:bobzzz@freenet.de
UID:cac4fef4-6351-4854-bbe4-9b6df857eaed
END:VCARD
"
.to_string(),
)
.unwrap();
assert_eq!(contacts[0].addr, "alice.mueller@posteo.de".to_string());
assert_eq!(contacts[0].display_name, "Alice Mueller".to_string());
assert_eq!(contacts[0].key, "".to_string());
assert_eq!(contacts[0].profile_photo, "".to_string());
assert_eq!(contacts[1].addr, "bobzzz@freenet.de".to_string());
assert_eq!(contacts[1].display_name, "".to_string());
assert_eq!(contacts[1].key, "".to_string());
assert_eq!(contacts[1].profile_photo, "".to_string());
assert_eq!(contacts.len(), 2);
}
#[test]
fn test_simple_example() {
let contacts = parse_vcard(
"BEGIN:VCARD
VERSION:4.0
FN:Alice Wonderland
N:Wonderland;Alice;;;Ms.
GENDER:W
EMAIL;TYPE=work:alice@example.com
KEY;TYPE=PGP;ENCODING=b:[base64-data]
REV:20240418T184242Z
END:VCARD"
.to_string(),
)
.unwrap();
assert_eq!(contacts[0].addr, "alice@example.com".to_string());
assert_eq!(contacts[0].display_name, "Alice Wonderland".to_string());
assert_eq!(contacts[0].key, "[base64-data]".to_string());
assert_eq!(contacts[0].profile_photo, "".to_string());
assert_eq!(contacts[0].timestamp, 1713465762); // I did not check whether this timestamp is correct
assert_eq!(contacts.len(), 1);
}
#[test]
fn test_contact_address() -> Result<()> {
let alice_addr = "alice@example.org";
@@ -277,4 +416,38 @@ mod tests {
assert!(EmailAddress::new("u@tt").is_ok());
assert_eq!(EmailAddress::new("@d.tt").is_ok(), false);
}
#[test]
fn test_android_contact_export() {
let contacts = parse_vcard(
"BEGIN:VCARD
VERSION:2.1
N:;Bob;;;
FN:Bob
TEL;CELL:+1-234-567-890
EMAIL;HOME:bob@example.org
END:VCARD
BEGIN:VCARD
VERSION:2.1
N:;Alice;;;
FN:Alice
EMAIL;HOME:alice@example.org
END:VCARD
"
.to_string(),
)
.unwrap();
assert_eq!(contacts[0].addr, "bob@example.org".to_string());
assert_eq!(contacts[0].display_name, "Bob".to_string());
assert_eq!(contacts[0].key, "".to_string());
assert_eq!(contacts[0].profile_photo, "".to_string());
assert_eq!(contacts[1].addr, "alice@example.org".to_string());
assert_eq!(contacts[1].display_name, "Alice".to_string());
assert_eq!(contacts[1].key, "".to_string());
assert_eq!(contacts[1].profile_photo, "".to_string());
assert_eq!(contacts.len(), 2);
}
}