Add new get_many_to_many XRPC endpoint #7

+1070 -16

11 changed files

Diff round #5

expand all

unified split

Cargo.lock

constellation

Cargo.toml

src

lib.rs

server

mod.rs

storage

mem_store.rs

mod.rs

rocks_store.rs

templates

get-many-to-many.html.j2

hello.html.j2

try-it-macros.html.j2

lexicons

blue.microcosm

links

getManyToMany.json

Cargo.lock

··· 1058 1058 "axum", 1059 1059 "axum-extra", 1060 1060 "axum-metrics", 1061 + "base64 0.22.1", 1061 1062 "bincode 1.3.3", 1062 1063 "clap", 1063 1064 "ctrlc",

constellation/Cargo.toml

··· 10 10 axum = "0.8.1" 11 11 axum-extra = { version = "0.10.0", features = ["query", "typed-header"] } 12 12 axum-metrics = "0.2" 13 + base64 = "0.22.1" 13 14 bincode = "1.3.3" 14 15 clap = { workspace = true } 15 16 ctrlc = "3.4.5"

+2 -2

constellation/src/lib.rs

··· 22 22 DeleteAccount(Did), 23 23 } 24 24 25 - #[derive(Debug, PartialEq, Eq, Hash, Clone, Serialize, Deserialize)] 25 + #[derive(Debug, Hash, Clone, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)] 26 26 pub struct Did(pub String); 27 27 28 28 impl<T: Into<String>> From<T> for Did { ··· 31 31 } 32 32 } 33 33 34 - #[derive(Debug, PartialEq, Serialize, Deserialize)] 34 + #[derive(Debug, Serialize, Deserialize, Clone, PartialEq, Eq, PartialOrd, Ord)] 35 35 pub struct RecordId { 36 36 pub did: Did, 37 37 pub collection: String,

+116

constellation/src/server/mod.rs

··· 114 114 }), 115 115 ) 116 116 .route( 117 + "/xrpc/blue.microcosm.links.getManyToMany", 118 + get({ 119 + let store = store.clone(); 120 + move |accept, query| async { 121 + spawn_blocking(|| get_many_to_many(accept, query, store)) 122 + .await 123 + .map_err(to500)? 124 + } 125 + }), 126 + ) 127 + .route( 117 128 "/xrpc/blue.microcosm.links.getBacklinks", 118 129 get({ 119 130 let store = store.clone(); ··· 661 672 GetLinkItemsResponse { 662 673 total: paged.total, 663 674 linking_records: paged.items, 675 + cursor, 676 + query: (*query).clone(), 677 + }, 678 + )) 679 + } 680 + 681 + #[derive(Clone, Deserialize)] 682 + #[serde(rename_all = "camelCase")] 683 + struct GetManyToManyItemsQuery { 684 + subject: String, 685 + source: String, 686 + /// path to the secondary link in the linking record 687 + path_to_other: String, 688 + /// filter to linking records (join of the m2m) by these DIDs 689 + #[serde(default)] 690 + did: Vec<String>, 691 + /// filter to specific secondary records 692 + #[serde(default)] 693 + other_subject: Vec<String>, 694 + cursor: Option<OpaqueApiCursor>, 695 + #[serde(default = "get_default_cursor_limit")] 696 + limit: u64, 697 + } 698 + #[derive(Debug, Serialize, Clone)] 699 + struct ManyToManyItem { 700 + link: RecordId, 701 + subject: String, 702 + } 703 + #[derive(Template, Serialize)] 704 + #[template(path = "get-many-to-many.html.j2")] 705 + struct GetManyToManyItemsResponse { 706 + items: Vec<ManyToManyItem>, 707 + cursor: Option<OpaqueApiCursor>, 708 + #[serde(skip_serializing)] 709 + query: GetManyToManyItemsQuery, 710 + } 711 + fn get_many_to_many( 712 + accept: ExtractAccept, 713 + query: axum_extra::extract::Query<GetManyToManyItemsQuery>, // supports multiple param occurrences 714 + store: impl LinkReader, 715 + ) -> Result<impl IntoResponse, http::StatusCode> { 716 + let after = query 717 + .cursor 718 + .clone() 719 + .map(|oc| ApiKeyedCursor::try_from(oc).map_err(|_| http::StatusCode::BAD_REQUEST)) 720 + .transpose()? 721 + .map(|c| c.next); 722 + 723 + let limit = query.limit; 724 + if limit > DEFAULT_CURSOR_LIMIT_MAX { 725 + return Err(http::StatusCode::BAD_REQUEST); 726 + } 727 + 728 + let filter_dids: HashSet<Did> = HashSet::from_iter( 729 + query 730 + .did 731 + .iter() 732 + .map(|d| d.trim()) 733 + .filter(|d| !d.is_empty()) 734 + .map(|d| Did(d.to_string())), 735 + ); 736 + 737 + let filter_other_subjects: HashSet<String> = HashSet::from_iter( 738 + query 739 + .other_subject 740 + .iter() 741 + .map(|s| s.trim().to_string()) 742 + .filter(|s| !s.is_empty()), 743 + ); 744 + 745 + let Some((collection, path)) = query.source.split_once(':') else { 746 + return Err(http::StatusCode::BAD_REQUEST); 747 + }; 748 + let path = format!(".{path}"); 749 + 750 + let path_to_other = format!(".{}", query.path_to_other); 751 + 752 + let paged = store 753 + .get_many_to_many( 754 + &query.subject, 755 + collection, 756 + &path, 757 + &path_to_other, 758 + limit, 759 + after, 760 + &filter_dids, 761 + &filter_other_subjects, 762 + ) 763 + .map_err(|_| http::StatusCode::INTERNAL_SERVER_ERROR)?; 764 + 765 + let cursor = paged.next.map(|next| ApiKeyedCursor { next }.into()); 766 + 767 + let items: Vec<ManyToManyItem> = paged 768 + .items 769 + .into_iter() 770 + .map(|(record_id, subject)| ManyToManyItem { 771 + link: record_id, 772 + subject, 773 + }) 774 + .collect(); 775 + 776 + Ok(acceptable( 777 + accept, 778 + GetManyToManyItemsResponse { 779 + items, 664 780 cursor, 665 781 query: (*query).clone(), 666 782 },

+129 -2

constellation/src/storage/mem_store.rs

··· 1 1 use super::{ 2 2 LinkReader, LinkStorage, Order, PagedAppendingCollection, PagedOrderedCollection, StorageStats, 3 3 }; 4 + use crate::storage::{decode_m2m_cursor, encode_m2m_cursor}; 4 5 use crate::{ActionableEvent, CountsByCount, Did, RecordId}; 5 6 use anyhow::Result; 6 7 use links::CollectedLink; ··· 197 198 items = items 198 199 .into_iter() 199 200 .skip_while(|(t, _, _)| after.as_ref().map(|a| t <= a).unwrap_or(false)) 200 - .take(limit as usize) 201 + .take(limit as usize + 1) 201 202 .collect(); 202 - let next = if items.len() as u64 >= limit { 203 + let next = if items.len() as u64 > limit { 204 + items.truncate(limit as usize); 203 205 items.last().map(|(t, _, _)| t.clone()) 204 206 } else { 205 207 None ··· 232 234 .map(|(did, _)| did) 233 235 .collect::<HashSet<_>>() 234 236 .len() as u64) 237 + } 238 + 239 + fn get_many_to_many( 240 + &self, 241 + target: &str, 242 + collection: &str, 243 + path: &str, 244 + path_to_other: &str, 245 + limit: u64, 246 + after: Option<String>, 247 + filter_dids: &HashSet<Did>, 248 + filter_to_targets: &HashSet<String>, 249 + ) -> Result<PagedOrderedCollection<(RecordId, String), String>> { 250 + let empty_res = Ok(PagedOrderedCollection { 251 + items: Vec::new(), 252 + next: None, 253 + }); 254 + 255 + let data = self.0.lock().unwrap(); 256 + 257 + let Some(sources) = data.targets.get(&Target::new(target)) else { 258 + return empty_res; 259 + }; 260 + let Some(linkers) = sources.get(&Source::new(collection, path)) else { 261 + return empty_res; 262 + }; 263 + let path_to_other = RecordPath::new(path_to_other); 264 + 265 + // Convert filter_to_targets to Target objects for comparison 266 + let filter_to_target_objs: HashSet<Target> = 267 + HashSet::from_iter(filter_to_targets.iter().map(|s| Target::new(s))); 268 + 269 + let mut grouped_links: HashMap<Target, Vec<RecordId>> = HashMap::new(); 270 + for (did, rkey) in linkers.iter().flatten().cloned() { 271 + // Filter by DID if filter is provided 272 + if !filter_dids.is_empty() && !filter_dids.contains(&did) { 273 + continue; 274 + } 275 + if let Some(fwd_target) = data 276 + .links 277 + .get(&did) 278 + .unwrap_or(&HashMap::new()) 279 + .get(&RepoId { 280 + collection: collection.to_string(), 281 + rkey: rkey.clone(), 282 + }) 283 + .unwrap_or(&Vec::new()) 284 + .iter() 285 + .find_map(|(path, target)| { 286 + if *path == path_to_other 287 + && (filter_to_target_objs.is_empty() 288 + || filter_to_target_objs.contains(target)) 289 + { 290 + Some(target) 291 + } else { 292 + None 293 + } 294 + }) 295 + { 296 + let record_ids = grouped_links.entry(fwd_target.clone()).or_default(); 297 + record_ids.push(RecordId { 298 + did, 299 + collection: collection.to_string(), 300 + rkey: rkey.0, 301 + }); 302 + } 303 + } 304 + 305 + let mut items = grouped_links 306 + .into_iter() 307 + .flat_map(|(target, records)| { 308 + records 309 + .iter() 310 + .map(move |r| (r.clone(), target.0.clone())) 311 + .collect::<Vec<_>>() 312 + }) 313 + .collect::<Vec<_>>(); 314 + 315 + // first try to sort by subject, then by did, collection and finally rkey 316 + items.sort_by(|a, b| { 317 + if a.1 == b.1 { 318 + a.0.cmp(&b.0) 319 + } else { 320 + a.1.cmp(&b.1) 321 + } 322 + }); 323 + 324 + // Parse cursor if provided (malformed cursor silently ignored) 325 + let after_cursor = after.and_then(|a| decode_m2m_cursor(&a).ok()); 326 + 327 + // Apply cursor: skip everything up to and including the cursor position 328 + items = items 329 + .into_iter() 330 + .skip_while(|item| { 331 + let Some((after_did, after_rkey, after_subject)) = &after_cursor else { 332 + return false; 333 + }; 334 + 335 + if &item.1 == after_subject { 336 + // Same subject — compare by RecordId to find our position 337 + let cursor_id = RecordId { 338 + did: Did(after_did.clone()), 339 + collection: collection.to_string(), 340 + rkey: after_rkey.clone(), 341 + }; 342 + item.0.cmp(&cursor_id).is_le() 343 + } else { 344 + // Different subject — compare subjects directly 345 + item.1.cmp(after_subject).is_le() 346 + } 347 + }) 348 + .take(limit as usize + 1) 349 + .collect(); 350 + 351 + // Build the new cursor from last item, if needed 352 + let next = if items.len() as u64 > limit { 353 + items.truncate(limit as usize); 354 + items 355 + .last() 356 + .map(|item| encode_m2m_cursor(&item.0.did.0, &item.0.rkey, &item.1)) 357 + } else { 358 + None 359 + }; 360 + 361 + Ok(PagedOrderedCollection { items, next }) 235 362 } 236 363 237 364 fn get_links(

+413

constellation/src/storage/mod.rs

··· 6 6 pub mod mem_store; 7 7 pub use mem_store::MemStorage; 8 8 9 + use anyhow::anyhow; 10 + 11 + use base64::engine::general_purpose as b64; 12 + use base64::Engine as _; 13 + 9 14 #[cfg(feature = "rocks")] 10 15 pub mod rocks_store; 11 16 #[cfg(feature = "rocks")] ··· 135 140 fn get_all_record_counts(&self, _target: &str) 136 141 -> Result<HashMap<String, HashMap<String, u64>>>; 137 142 143 + fn get_many_to_many( 144 + &self, 145 + target: &str, 146 + collection: &str, 147 + path: &str, 148 + path_to_other: &str, 149 + limit: u64, 150 + after: Option<String>, 151 + filter_dids: &HashSet<Did>, 152 + filter_to_targets: &HashSet<String>, 153 + ) -> Result<PagedOrderedCollection<(RecordId, String), String>>; 154 + 138 155 fn get_all_counts( 139 156 &self, 140 157 _target: &str, ··· 142 159 143 160 /// assume all stats are estimates, since exact counts are very challenging for LSMs 144 161 fn get_stats(&self) -> Result<StorageStats>; 162 + } 163 + 164 + // Shared helpers 165 + 166 + /// Decode a base64 cursor into its component parts (did, rkey, subject). 167 + /// The subject is placed last because it may contain '|' characters. 168 + pub(crate) fn decode_m2m_cursor(cursor: &str) -> Result<(String, String, String)> { 169 + let decoded = String::from_utf8(b64::URL_SAFE.decode(cursor)?)?; 170 + let mut parts = decoded.splitn(3, '|').map(String::from); 171 + 172 + // Using .next() to pull each part out of the iterator in order. 173 + // This avoids collecting into a Vec just to index and clone back out. 174 + let did = parts 175 + .next() 176 + .ok_or_else(|| anyhow!("missing did in cursor"))?; 177 + let rkey = parts 178 + .next() 179 + .ok_or_else(|| anyhow!("missing rkey in cursor"))?; 180 + let subject = parts 181 + .next() 182 + .ok_or_else(|| anyhow!("missing subject in cursor"))?; 183 + 184 + Ok((did, rkey, subject)) 185 + } 186 + 187 + /// Encode cursor components into a base64 string. 188 + pub(crate) fn encode_m2m_cursor(did: &str, rkey: &str, subject: &str) -> String { 189 + let raw = format!("{did}|{rkey}|{subject}"); 190 + b64::URL_SAFE.encode(&raw) 145 191 } 146 192 147 193 #[cfg(test)] ··· 1669 1715 items: vec![("b.com".to_string(), 2, 2),], 1670 1716 next: None, 1671 1717 } 1718 + ); 1719 + 1720 + // Pagination edge cases: we have 2 grouped results (b.com and c.com) 1721 + 1722 + // Case 1: limit > items (limit=10, items=2) -> next should be None 1723 + let result = storage.get_many_to_many_counts( 1724 + "a.com", 1725 + "app.t.c", 1726 + ".abc.uri", 1727 + ".def.uri", 1728 + 10, 1729 + None, 1730 + &HashSet::new(), 1731 + &HashSet::new(), 1732 + )?; 1733 + assert_eq!(result.items.len(), 2); 1734 + assert_eq!(result.next, None, "next should be None when items < limit"); 1735 + 1736 + // Case 2: limit == items (limit=2, items=2) -> next should be None 1737 + let result = storage.get_many_to_many_counts( 1738 + "a.com", 1739 + "app.t.c", 1740 + ".abc.uri", 1741 + ".def.uri", 1742 + 2, 1743 + None, 1744 + &HashSet::new(), 1745 + &HashSet::new(), 1746 + )?; 1747 + assert_eq!(result.items.len(), 2); 1748 + assert_eq!( 1749 + result.next, None, 1750 + "next should be None when items == limit (no more pages)" 1751 + ); 1752 + 1753 + // Case 3: limit < items (limit=1, items=2) -> next should be Some 1754 + let result = storage.get_many_to_many_counts( 1755 + "a.com", 1756 + "app.t.c", 1757 + ".abc.uri", 1758 + ".def.uri", 1759 + 1, 1760 + None, 1761 + &HashSet::new(), 1762 + &HashSet::new(), 1763 + )?; 1764 + assert_eq!(result.items.len(), 1); 1765 + assert!( 1766 + result.next.is_some(), 1767 + "next should be Some when items > limit" 1768 + ); 1769 + 1770 + // Verify second page returns remaining item with no cursor 1771 + let result2 = storage.get_many_to_many_counts( 1772 + "a.com", 1773 + "app.t.c", 1774 + ".abc.uri", 1775 + ".def.uri", 1776 + 1, 1777 + result.next, 1778 + &HashSet::new(), 1779 + &HashSet::new(), 1780 + )?; 1781 + assert_eq!(result2.items.len(), 1); 1782 + assert_eq!(result2.next, None, "next should be None on final page"); 1783 + }); 1784 + 1785 + test_each_storage!(get_m2m_empty, |storage| { 1786 + assert_eq!( 1787 + storage.get_many_to_many( 1788 + "a.com", 1789 + "a.b.c", 1790 + ".d.e", 1791 + ".f.g", 1792 + 10, 1793 + None, 1794 + &HashSet::new(), 1795 + &HashSet::new(), 1796 + )?, 1797 + PagedOrderedCollection { 1798 + items: vec![], 1799 + next: None, 1800 + } 1801 + ); 1802 + }); 1803 + 1804 + test_each_storage!(get_m2m_single, |storage| { 1805 + storage.push( 1806 + &ActionableEvent::CreateLinks { 1807 + record_id: RecordId { 1808 + did: "did:plc:asdf".into(), 1809 + collection: "app.t.c".into(), 1810 + rkey: "asdf".into(), 1811 + }, 1812 + links: vec![ 1813 + CollectedLink { 1814 + target: Link::Uri("a.com".into()), 1815 + path: ".abc.uri".into(), 1816 + }, 1817 + CollectedLink { 1818 + target: Link::Uri("b.com".into()), 1819 + path: ".def.uri".into(), 1820 + }, 1821 + CollectedLink { 1822 + target: Link::Uri("b.com".into()), 1823 + path: ".ghi.uri".into(), 1824 + }, 1825 + ], 1826 + }, 1827 + 0, 1828 + )?; 1829 + assert_eq!( 1830 + storage.get_many_to_many( 1831 + "a.com", 1832 + "app.t.c", 1833 + ".abc.uri", 1834 + ".def.uri", 1835 + 10, 1836 + None, 1837 + &HashSet::new(), 1838 + &HashSet::new(), 1839 + )?, 1840 + PagedOrderedCollection { 1841 + items: vec![( 1842 + RecordId { 1843 + did: "did:plc:asdf".into(), 1844 + collection: "app.t.c".into(), 1845 + rkey: "asdf".into(), 1846 + }, 1847 + "b.com".to_string(), 1848 + )], 1849 + next: None, 1850 + } 1851 + ); 1852 + }); 1853 + 1854 + test_each_storage!(get_m2m_filters, |storage| { 1855 + storage.push( 1856 + &ActionableEvent::CreateLinks { 1857 + record_id: RecordId { 1858 + did: "did:plc:asdf".into(), 1859 + collection: "app.t.c".into(), 1860 + rkey: "asdf".into(), 1861 + }, 1862 + links: vec![ 1863 + CollectedLink { 1864 + target: Link::Uri("a.com".into()), 1865 + path: ".abc.uri".into(), 1866 + }, 1867 + CollectedLink { 1868 + target: Link::Uri("b.com".into()), 1869 + path: ".def.uri".into(), 1870 + }, 1871 + ], 1872 + }, 1873 + 0, 1874 + )?; 1875 + storage.push( 1876 + &ActionableEvent::CreateLinks { 1877 + record_id: RecordId { 1878 + did: "did:plc:asdf".into(), 1879 + collection: "app.t.c".into(), 1880 + rkey: "asdf2".into(), 1881 + }, 1882 + links: vec![ 1883 + CollectedLink { 1884 + target: Link::Uri("a.com".into()), 1885 + path: ".abc.uri".into(), 1886 + }, 1887 + CollectedLink { 1888 + target: Link::Uri("b.com".into()), 1889 + path: ".def.uri".into(), 1890 + }, 1891 + ], 1892 + }, 1893 + 1, 1894 + )?; 1895 + storage.push( 1896 + &ActionableEvent::CreateLinks { 1897 + record_id: RecordId { 1898 + did: "did:plc:fdsa".into(), 1899 + collection: "app.t.c".into(), 1900 + rkey: "fdsa".into(), 1901 + }, 1902 + links: vec![ 1903 + CollectedLink { 1904 + target: Link::Uri("a.com".into()), 1905 + path: ".abc.uri".into(), 1906 + }, 1907 + CollectedLink { 1908 + target: Link::Uri("c.com".into()), 1909 + path: ".def.uri".into(), 1910 + }, 1911 + ], 1912 + }, 1913 + 2, 1914 + )?; 1915 + storage.push( 1916 + &ActionableEvent::CreateLinks { 1917 + record_id: RecordId { 1918 + did: "did:plc:fdsa".into(), 1919 + collection: "app.t.c".into(), 1920 + rkey: "fdsa2".into(), 1921 + }, 1922 + links: vec![ 1923 + CollectedLink { 1924 + target: Link::Uri("a.com".into()), 1925 + path: ".abc.uri".into(), 1926 + }, 1927 + CollectedLink { 1928 + target: Link::Uri("c.com".into()), 1929 + path: ".def.uri".into(), 1930 + }, 1931 + ], 1932 + }, 1933 + 3, 1934 + )?; 1935 + 1936 + // Test without filters - should get all records as flat items 1937 + let result = storage.get_many_to_many( 1938 + "a.com", 1939 + "app.t.c", 1940 + ".abc.uri", 1941 + ".def.uri", 1942 + 10, 1943 + None, 1944 + &HashSet::new(), 1945 + &HashSet::new(), 1946 + )?; 1947 + assert_eq!(result.items.len(), 4); 1948 + assert_eq!(result.next, None); 1949 + // Check b.com items 1950 + let b_items: Vec<_> = result 1951 + .items 1952 + .iter() 1953 + .filter(|(_, subject)| subject == "b.com") 1954 + .collect(); 1955 + assert_eq!(b_items.len(), 2); 1956 + assert!(b_items 1957 + .iter() 1958 + .any(|(r, _)| r.did.0 == "did:plc:asdf" && r.rkey == "asdf")); 1959 + assert!(b_items 1960 + .iter() 1961 + .any(|(r, _)| r.did.0 == "did:plc:asdf" && r.rkey == "asdf2")); 1962 + // Check c.com items 1963 + let c_items: Vec<_> = result 1964 + .items 1965 + .iter() 1966 + .filter(|(_, subject)| subject == "c.com") 1967 + .collect(); 1968 + assert_eq!(c_items.len(), 2); 1969 + assert!(c_items 1970 + .iter() 1971 + .any(|(r, _)| r.did.0 == "did:plc:fdsa" && r.rkey == "fdsa")); 1972 + assert!(c_items 1973 + .iter() 1974 + .any(|(r, _)| r.did.0 == "did:plc:fdsa" && r.rkey == "fdsa2")); 1975 + 1976 + // Test with DID filter - should only get records from did:plc:fdsa 1977 + let result = storage.get_many_to_many( 1978 + "a.com", 1979 + "app.t.c", 1980 + ".abc.uri", 1981 + ".def.uri", 1982 + 10, 1983 + None, 1984 + &HashSet::from_iter([Did("did:plc:fdsa".to_string())]), 1985 + &HashSet::new(), 1986 + )?; 1987 + assert_eq!(result.items.len(), 2); 1988 + assert!(result.items.iter().all(|(_, subject)| subject == "c.com")); 1989 + assert!(result.items.iter().all(|(r, _)| r.did.0 == "did:plc:fdsa")); 1990 + 1991 + // Test with target filter - should only get records linking to b.com 1992 + let result = storage.get_many_to_many( 1993 + "a.com", 1994 + "app.t.c", 1995 + ".abc.uri", 1996 + ".def.uri", 1997 + 10, 1998 + None, 1999 + &HashSet::new(), 2000 + &HashSet::from_iter(["b.com".to_string()]), 2001 + )?; 2002 + assert_eq!(result.items.len(), 2); 2003 + assert!(result.items.iter().all(|(_, subject)| subject == "b.com")); 2004 + assert!(result.items.iter().all(|(r, _)| r.did.0 == "did:plc:asdf")); 2005 + 2006 + // Pagination edge cases: we have 4 flat items 2007 + 2008 + // Case 1: limit > items (limit=10, items=4) -> next should be None 2009 + let result = storage.get_many_to_many( 2010 + "a.com", 2011 + "app.t.c", 2012 + ".abc.uri", 2013 + ".def.uri", 2014 + 10, 2015 + None, 2016 + &HashSet::new(), 2017 + &HashSet::new(), 2018 + )?; 2019 + assert_eq!(result.items.len(), 4); 2020 + assert_eq!(result.next, None, "next should be None when items < limit"); 2021 + 2022 + // Case 2: limit == items (limit=4, items=4) -> next should be None 2023 + let result = storage.get_many_to_many( 2024 + "a.com", 2025 + "app.t.c", 2026 + ".abc.uri", 2027 + ".def.uri", 2028 + 4, 2029 + None, 2030 + &HashSet::new(), 2031 + &HashSet::new(), 2032 + )?; 2033 + assert_eq!(result.items.len(), 4); 2034 + assert_eq!( 2035 + result.next, None, 2036 + "next should be None when items == limit (no more pages)" 2037 + ); 2038 + 2039 + // Case 3: limit < items (limit=3, items=4) -> next should be Some 2040 + let result = storage.get_many_to_many( 2041 + "a.com", 2042 + "app.t.c", 2043 + ".abc.uri", 2044 + ".def.uri", 2045 + 3, 2046 + None, 2047 + &HashSet::new(), 2048 + &HashSet::new(), 2049 + )?; 2050 + assert_eq!(result.items.len(), 3); 2051 + assert!( 2052 + result.next.is_some(), 2053 + "next should be Some when items > limit" 2054 + ); 2055 + 2056 + // Verify second page returns remaining item with no cursor. 2057 + // This now works correctly because we use a composite cursor that includes 2058 + // (target, did, rkey), allowing pagination even when multiple records share 2059 + // the same target string. 2060 + let result2 = storage.get_many_to_many( 2061 + "a.com", 2062 + "app.t.c", 2063 + ".abc.uri", 2064 + ".def.uri", 2065 + 3, 2066 + result.next, 2067 + &HashSet::new(), 2068 + &HashSet::new(), 2069 + )?; 2070 + assert_eq!( 2071 + result2.items.len(), 2072 + 1, 2073 + "second page should have 1 remaining item" 2074 + ); 2075 + assert_eq!(result2.next, None, "next should be None on final page"); 2076 + 2077 + // Verify we got all 4 unique items across both pages (no duplicates, no gaps) 2078 + let mut all_rkeys: Vec<_> = result.items.iter().map(|(r, _)| r.rkey.clone()).collect(); 2079 + all_rkeys.extend(result2.items.iter().map(|(r, _)| r.rkey.clone())); 2080 + all_rkeys.sort(); 2081 + assert_eq!( 2082 + all_rkeys, 2083 + vec!["asdf", "asdf2", "fdsa", "fdsa2"], 2084 + "should have all 4 records across both pages" 1672 2085 ); 1673 2086 }); 1674 2087 }

+197 -12

constellation/src/storage/rocks_store.rs

··· 2 2 ActionableEvent, LinkReader, LinkStorage, Order, PagedAppendingCollection, 3 3 PagedOrderedCollection, StorageStats, 4 4 }; 5 + use crate::storage::{decode_m2m_cursor, encode_m2m_cursor}; 5 6 use crate::{CountsByCount, Did, RecordId}; 6 7 use anyhow::{bail, Result}; 7 8 use bincode::Options as BincodeOptions; ··· 1032 1033 1033 1034 // aand we can skip target ids that must be on future pages 1034 1035 // (this check continues after the did-lookup, which we have to do) 1035 - let page_is_full = grouped_counts.len() as u64 >= limit; 1036 + let page_is_full = grouped_counts.len() as u64 > limit; 1036 1037 if page_is_full { 1037 - let current_max = grouped_counts.keys().next_back().unwrap(); // limit should be non-zero bleh 1038 + let current_max = grouped_counts.keys().next_back().unwrap(); 1038 1039 if fwd_target > *current_max { 1039 1040 continue; 1040 1041 } ··· 1070 1071 } 1071 1072 } 1072 1073 1074 + // If we accumulated more than limit groups, there's another page. 1075 + // Pop the extra before building items so it doesn't appear in results. 1076 + let next = if grouped_counts.len() as u64 > limit { 1077 + grouped_counts.pop_last(); 1078 + grouped_counts 1079 + .keys() 1080 + .next_back() 1081 + .map(|k| format!("{}", k.0)) 1082 + } else { 1083 + None 1084 + }; 1085 + 1073 1086 let mut items: Vec<(String, u64, u64)> = Vec::with_capacity(grouped_counts.len()); 1074 1087 for (target_id, (n, dids)) in &grouped_counts { 1075 1088 let Some(target) = self ··· 1081 1094 }; 1082 1095 items.push((target.0 .0, *n, dids.len() as u64)); 1083 1096 } 1084 - 1085 - let next = if grouped_counts.len() as u64 >= limit { 1086 - // yeah.... it's a number saved as a string......sorry 1087 - grouped_counts 1088 - .keys() 1089 - .next_back() 1090 - .map(|k| format!("{}", k.0)) 1091 - } else { 1092 - None 1093 - }; 1094 1097 1095 1098 Ok(PagedOrderedCollection { items, next }) 1096 1099 } ··· 1120 1123 } else { 1121 1124 Ok(0) 1122 1125 } 1126 + } 1127 + 1128 + fn get_many_to_many( 1129 + &self, 1130 + target: &str, 1131 + collection: &str, 1132 + path: &str, 1133 + path_to_other: &str, 1134 + limit: u64, 1135 + after: Option<String>, 1136 + filter_dids: &HashSet<Did>, 1137 + filter_to_targets: &HashSet<String>, 1138 + ) -> Result<PagedOrderedCollection<(RecordId, String), String>> { 1139 + let collection = Collection(collection.to_string()); 1140 + let path = RPath(path.to_string()); 1141 + 1142 + let target_key = TargetKey(Target(target.to_string()), collection.clone(), path); 1143 + 1144 + // Parse cursor if provided (malformed cursor silently ignored) 1145 + let after_cursor = after.and_then(|a| decode_m2m_cursor(&a).ok()); 1146 + 1147 + let Some(target_id) = self.target_id_table.get_id_val(&self.db, &target_key)? else { 1148 + eprintln!("Target not found for {target_key:?}"); 1149 + return Ok(PagedOrderedCollection::empty()); 1150 + }; 1151 + 1152 + let filter_did_ids: HashMap<DidId, bool> = filter_dids 1153 + .iter() 1154 + .filter_map(|did| self.did_id_table.get_id_val(&self.db, did).transpose()) 1155 + .collect::<Result<Vec<DidIdValue>>>()? 1156 + .into_iter() 1157 + .map(|DidIdValue(id, active)| (id, active)) 1158 + .collect(); 1159 + 1160 + let mut filter_to_target_ids: HashSet<TargetId> = HashSet::new(); 1161 + for t in filter_to_targets { 1162 + for (_, target_id) in self.iter_targets_for_target(&Target(t.to_string())) { 1163 + filter_to_target_ids.insert(target_id); 1164 + } 1165 + } 1166 + 1167 + let linkers = self.get_target_linkers(&target_id)?; 1168 + 1169 + // we want to provide many to many which effectively means that we want to show a specific 1170 + // list of reords that is linked to by a specific number of linkers 1171 + let mut grouped_links: BTreeMap<TargetId, Vec<RecordId>> = BTreeMap::new(); 1172 + for (did_id, rkey) in linkers.0 { 1173 + if did_id.is_empty() { 1174 + continue; 1175 + } 1176 + 1177 + if !filter_did_ids.is_empty() && filter_did_ids.get(&did_id) != Some(&true) { 1178 + continue; 1179 + } 1180 + 1181 + // Make sure the current did is active 1182 + let Some(did) = self.did_id_table.get_val_from_id(&self.db, did_id.0)? else { 1183 + eprintln!("failed to look up did from did_id {did_id:?}"); 1184 + continue; 1185 + }; 1186 + let Some(DidIdValue(_, active)) = self.did_id_table.get_id_val(&self.db, &did)? else { 1187 + eprintln!("failed to look up did_value from did_id {did_id:?}: {did:?}: data consistency bug?"); 1188 + continue; 1189 + }; 1190 + if !active { 1191 + continue; 1192 + } 1193 + 1194 + let record_link_key = RecordLinkKey(did_id, collection.clone(), rkey.clone()); 1195 + let Some(targets) = self.get_record_link_targets(&record_link_key)? else { 1196 + continue; 1197 + }; 1198 + 1199 + let Some(fwd_target) = targets 1200 + .0 1201 + .into_iter() 1202 + .filter_map(|RecordLinkTarget(rpath, target_id)| { 1203 + if rpath.0 == path_to_other 1204 + && (filter_to_target_ids.is_empty() 1205 + || filter_to_target_ids.contains(&target_id)) 1206 + { 1207 + Some(target_id) 1208 + } else { 1209 + None 1210 + } 1211 + }) 1212 + .take(1) 1213 + .next() 1214 + else { 1215 + eprintln!("no forward match found."); 1216 + continue; 1217 + }; 1218 + 1219 + let page_is_full = grouped_links.len() as u64 >= limit; 1220 + if page_is_full { 1221 + let current_max = grouped_links.keys().next_back().unwrap(); 1222 + if fwd_target > *current_max { 1223 + continue; 1224 + } 1225 + } 1226 + 1227 + // link to be added 1228 + let record_id = RecordId { 1229 + did, 1230 + collection: collection.0.clone(), 1231 + rkey: rkey.0, 1232 + }; 1233 + 1234 + // pagination: 1235 + if after_cursor.is_some() { 1236 + // extract composite-cursor parts 1237 + let Some((after_did, after_rkey, after_subject)) = &after_cursor else { 1238 + continue; 1239 + }; 1240 + 1241 + let Some(fwd_target_key) = self 1242 + .target_id_table 1243 + .get_val_from_id(&self.db, fwd_target.0)? 1244 + else { 1245 + eprintln!("failed to look up target from target_id {fwd_target:?}"); 1246 + continue; 1247 + }; 1248 + 1249 + // first try and compare by subject only 1250 + if &fwd_target_key.0 .0 != after_subject 1251 + && fwd_target_key.0 .0.cmp(after_subject).is_le() 1252 + { 1253 + continue; 1254 + } 1255 + 1256 + // then, if needed, we compare by record id 1257 + let cursor_id = RecordId { 1258 + did: Did(after_did.clone()), 1259 + collection: collection.0.clone(), 1260 + rkey: after_rkey.clone(), 1261 + }; 1262 + if record_id.cmp(&cursor_id).is_le() { 1263 + continue; 1264 + } 1265 + } 1266 + 1267 + // pagination, continued 1268 + let mut should_evict = false; 1269 + let entry = grouped_links.entry(fwd_target.clone()).or_insert_with(|| { 1270 + should_evict = page_is_full; 1271 + Vec::default() 1272 + }); 1273 + entry.push(record_id); 1274 + 1275 + if should_evict { 1276 + grouped_links.pop_last(); 1277 + } 1278 + } 1279 + 1280 + let mut items: Vec<(RecordId, String)> = Vec::with_capacity(grouped_links.len()); 1281 + for (fwd_target_id, records) in &grouped_links { 1282 + let Some(target_key) = self 1283 + .target_id_table 1284 + .get_val_from_id(&self.db, fwd_target_id.0)? 1285 + else { 1286 + eprintln!("failed to look up target from target_id {fwd_target_id:?}"); 1287 + continue; 1288 + }; 1289 + 1290 + let target_string = target_key.0 .0; 1291 + 1292 + records 1293 + .iter() 1294 + .for_each(|r| items.push((r.clone(), target_string.clone()))); 1295 + } 1296 + 1297 + // Build new cursor from last the item, if needed 1298 + let next = if items.len() as u64 > limit { 1299 + items.truncate(limit as usize); 1300 + items 1301 + .last() 1302 + .map(|item| encode_m2m_cursor(&item.0.did.0, &item.0.rkey, &item.1)) 1303 + } else { 1304 + None 1305 + }; 1306 + 1307 + Ok(PagedOrderedCollection { items, next }) 1123 1308 } 1124 1309 1125 1310 fn get_links(

+58

constellation/templates/get-many-to-many.html.j2

··· 1 + {% extends "base.html.j2" %} 2 + {% import "try-it-macros.html.j2" as try_it %} 3 + 4 + {% block title %}Many-to-Many Links{% endblock %} 5 + {% block description %}All {{ query.source }} records with many-to-many links to {{ query.subject }} joining through {{ query.path_to_other }}{% endblock %} 6 + 7 + {% block content %} 8 + 9 + {% call try_it::get_many_to_many(query.subject, query.source, query.path_to_other, query.did, query.other_subject, query.limit) %} 10 + 11 + <h2> 12 + Many-to-many links to <code>{{ query.subject }}</code> 13 + {% if let Some(browseable_uri) = query.subject|to_browseable %} 14 + <small style="font-weight: normal; font-size: 1rem"><a href="{{ browseable_uri }}">browse record</a></small> 15 + {% endif %} 16 + </h2> 17 + 18 + <p><strong>Many-to-many links</strong> from <code>{{ query.source }}</code> joining through <code>{{ query.path_to_other }}</code></p> 19 + 20 + <ul> 21 + <li>See all links to this target at <code>/links/all</code>: <a href="/links/all?target={{ query.subject|urlencode }}">/links/all?target={{ query.subject }}</a></li> 22 + </ul> 23 + 24 + <h3>Many-to-many links, most recent first:</h3> 25 + 26 + {% for item in items %} 27 + <pre style="display: block; margin: 1em 2em" class="code"><strong>Subject</strong>: <a href="/links/all?target={{ item.subject|urlencode }}">{{ item.subject }}</a> 28 + <strong>DID</strong>: {{ item.link.did().0 }} 29 + <strong>Collection</strong>: {{ item.link.collection }} 30 + <strong>RKey</strong>: {{ item.link.rkey }} 31 + -> <a href="https://pdsls.dev/at://{{ item.link.did().0 }}/{{ item.link.collection }}/{{ item.link.rkey }}">browse record</a></pre> 32 + {% endfor %} 33 + 34 + {% if let Some(c) = cursor %} 35 + <form method="get" action="/xrpc/blue.microcosm.links.getManyToMany"> 36 + <input type="hidden" name="subject" value="{{ query.subject }}" /> 37 + <input type="hidden" name="source" value="{{ query.source }}" /> 38 + <input type="hidden" name="pathToOther" value="{{ query.path_to_other }}" /> 39 + {% for did in query.did %} 40 + <input type="hidden" name="did" value="{{ did }}" /> 41 + {% endfor %} 42 + {% for other in query.other_subject %} 43 + <input type="hidden" name="otherSubject" value="{{ other }}" /> 44 + {% endfor %} 45 + <input type="hidden" name="limit" value="{{ query.limit }}" /> 46 + <input type="hidden" name="cursor" value={{ c|json|safe }} /> 47 + <button type="submit">next page…</button> 48 + </form> 49 + {% else %} 50 + <button disabled><em>end of results</em></button> 51 + {% endif %} 52 + 53 + <details> 54 + <summary>Raw JSON response</summary> 55 + <pre class="code">{{ self|tojson }}</pre> 56 + </details> 57 + 58 + {% endblock %}

+19

constellation/templates/hello.html.j2

··· 83 83 ) %} 84 84 85 85 86 + <h3 class="route"><code>GET /xrpc/blue.microcosm.links.getManyToMany</code></h3> 87 + 88 + <p>A list of many-to-many join records linking to a target and a secondary target.</p> 89 + 90 + <h4>Query parameters:</h4> 91 + 92 + <ul> 93 + <li><p><code>subject</code>: required, must url-encode. Example: <code>at://did:plc:vc7f4oafdgxsihk4cry2xpze/app.bsky.feed.post/3lgwdn7vd722r</code></p></li> 94 + <li><p><code>source</code>: required. Example: <code>app.bsky.feed.like:subject.uri</code></p></li> 95 + <li><p><code>pathToOther</code>: required. Path to the secondary link in the many-to-many record. Example: <code>otherThing.uri</code></p></li> 96 + <li><p><code>did</code>: optional, filter links to those from specific users. Include multiple times to filter by multiple users. Example: <code>did=did:plc:vc7f4oafdgxsihk4cry2xpze&did=did:plc:vc7f4oafdgxsihk4cry2xpze</code></p></li> 97 + <li><p><code>otherSubject</code>: optional, filter secondary links to specific subjects. Include multiple times to filter by multiple subjects. Example: <code>at://did:plc:vc7f4oafdgxsihk4cry2xpze/app.bsky.feed.post/3lgwdn7vd722r</code></p></li> 98 + <li><p><code>limit</code>: optional. Default: <code>16</code>. Maximum: <code>100</code></p></li> 99 + </ul> 100 + 101 + <p style="margin-bottom: 0"><strong>Try it:</strong></p> 102 + {% call try_it::get_many_to_many("at://did:plc:a4pqq234yw7fqbddawjo7y35/app.bsky.feed.post/3m237ilwc372e", "app.bsky.feed.like:subject.uri", "reply.parent.uri", [""], [""], 16) %} 103 + 104 + 86 105 <h3 class="route"><code>GET /links</code></h3> 87 106 88 107 <p>A list of records linking to a target.</p>

+30

constellation/templates/try-it-macros.html.j2

··· 68 68 </script> 69 69 {% endmacro %} 70 70 71 + {% macro get_many_to_many(subject, source, pathToOther, dids, otherSubjects, limit) %} 72 + <form method="get" action="/xrpc/blue.microcosm.links.getManyToMany"> 73 + <pre class="code"><strong>GET</strong> /xrpc/blue.microcosm.links.getManyToMany 74 + ?subject= <input type="text" name="subject" value="{{ subject }}" placeholder="at-uri, did, uri..." /> 75 + &source= <input type="text" name="source" value="{{ source }}" placeholder="app.bsky.feed.like:subject" /> 76 + &pathToOther= <input type="text" name="pathToOther" value="{{ pathToOther }}" placeholder="otherThing" /> 77 + {%- for did in dids %}{% if !did.is_empty() %} 78 + &did= <input type="text" name="did" value="{{ did }}" placeholder="did:plc:..." />{% endif %}{% endfor %} 79 + <span id="m2m-did-placeholder"></span> <button id="m2m-add-did">+ did filter</button> 80 + {%- for otherSubject in otherSubjects %}{% if !otherSubject.is_empty() %} 81 + &otherSubject= <input type="text" name="otherSubject" value="{{ otherSubject }}" placeholder="at-uri, did, uri..." />{% endif %}{% endfor %} 82 + <span id="m2m-other-placeholder"></span> <button id="m2m-add-other">+ other subject filter</button> 83 + &limit= <input type="number" name="limit" value="{{ limit }}" max="100" placeholder="100" /> <button type="submit">get many-to-many links</button></pre> 84 + </form> 85 + <script> 86 + const m2mAddDidButton = document.getElementById('m2m-add-did'); 87 + const m2mDidPlaceholder = document.getElementById('m2m-did-placeholder'); 88 + m2mAddDidButton.addEventListener('click', e => { 89 + e.preventDefault(); 90 + const i = document.createElement('input'); 91 + i.placeholder = 'did:plc:...'; 92 + i.name = "did" 93 + const p = m2mAddDidButton.parentNode; 94 + p.insertBefore(document.createTextNode('&did= '), m2mDidPlaceholder); 95 + p.insertBefore(i, m2mDidPlaceholder); 96 + p.insertBefore(document.createTextNode('\n '), m2mDidPlaceholder); 97 + }); 98 + </script> 99 + {% endmacro %} 100 + 71 101 {% macro links(target, collection, path, dids, limit) %} 72 102 <form method="get" action="/links"> 73 103 <pre class="code"><strong>GET</strong> /links

+104

lexicons/blue.microcosm/links/getManyToMany.json

··· 1 + { 2 + "lexicon": 1, 3 + "id": "blue.microcosm.links.getManyToMany", 4 + "defs": { 5 + "main": { 6 + "type": "query", 7 + "description": "Get records that link to a primary subject along with the secondary subjects they also reference", 8 + "parameters": { 9 + "type": "params", 10 + "required": ["subject", "source", "pathToOther"], 11 + "properties": { 12 + "subject": { 13 + "type": "string", 14 + "format": "uri", 15 + "description": "the primary target being linked to (at-uri, did, or uri)" 16 + }, 17 + "source": { 18 + "type": "string", 19 + "description": "collection and path specification for the primary link (e.g., 'app.bsky.feed.like:subject.uri')" 20 + }, 21 + "pathToOther": { 22 + "type": "string", 23 + "description": "path to the secondary link in the many-to-many record (e.g., 'otherThing.uri')" 24 + }, 25 + "did": { 26 + "type": "array", 27 + "description": "filter links to those from specific users", 28 + "items": { 29 + "type": "string", 30 + "format": "did" 31 + } 32 + }, 33 + "otherSubject": { 34 + "type": "array", 35 + "description": "filter secondary links to specific subjects", 36 + "items": { 37 + "type": "string" 38 + } 39 + }, 40 + "limit": { 41 + "type": "integer", 42 + "minimum": 1, 43 + "maximum": 100, 44 + "default": 16, 45 + "description": "number of results to return" 46 + } 47 + } 48 + }, 49 + "output": { 50 + "encoding": "application/json", 51 + "schema": { 52 + "type": "object", 53 + "required": ["items"], 54 + "properties": { 55 + "items": { 56 + "type": "array", 57 + "items": { 58 + "type": "ref", 59 + "ref": "#item" 60 + } 61 + }, 62 + "cursor": { 63 + "type": "string" 64 + } 65 + } 66 + } 67 + } 68 + }, 69 + "item": { 70 + "type": "object", 71 + "required": ["link", "subject"], 72 + "properties": { 73 + "link": { 74 + "type": "ref", 75 + "ref": "#linkRecord" 76 + }, 77 + "subject": { 78 + "type": "string" 79 + } 80 + } 81 + }, 82 + "linkRecord": { 83 + "type": "object", 84 + "required": ["did", "collection", "rkey"], 85 + "description": "A record identifier consisting of a DID, collection, and record key", 86 + "properties": { 87 + "did": { 88 + "type": "string", 89 + "format": "did", 90 + "description": "the DID of the linking record's repository" 91 + }, 92 + "collection": { 93 + "type": "string", 94 + "format": "nsid", 95 + "description": "the collection of the linking record" 96 + }, 97 + "rkey": { 98 + "type": "string", 99 + "format": "record-key" 100 + } 101 + } 102 + } 103 + } 104 + }

History

8 rounds 13 comments

seoul.systems submitted #7 3w

diff interdiff

11 commits

expand

45e798b8

wip: m2m

f3144f8d

Add tests for new get_many_to_many query handler

21057eca

Fix get_m2m_empty test

4d686c35

Replace tuple with RecordsBySubject struct

4e358f1a

Fix conflicts after rebasing on main

bb107c03

Use record_id/subject tuple as return type for get_many_to_many

90220d0b

Fix get_many_to_many pagination with composite cursor

373d0744

Fix get_many_to_many_counts pagination with fetch N+1

46f5fae4

wip

6d22f695

Fix rocks-store to match mem-store composite cursor

c2d01284

Address feedback from fig

expand 0 comments

pull request successfully merged

seoul.systems submitted #6 4w

diff interdiff

10 commits

expand

45e798b8

wip: m2m

f3144f8d

Add tests for new get_many_to_many query handler

21057eca

Fix get_m2m_empty test

4d686c35

Replace tuple with RecordsBySubject struct

4e358f1a

Fix conflicts after rebasing on main

bb107c03

Use record_id/subject tuple as return type for get_many_to_many

90220d0b

Fix get_many_to_many pagination with composite cursor

373d0744

Fix get_many_to_many_counts pagination with fetch N+1

46f5fae4

wip

6d22f695

Fix rocks-store to match mem-store composite cursor

expand 0 comments

seoul.systems submitted #5 4w

interdiff

8 commits

expand

45e798b8

wip: m2m

f3144f8d

Add tests for new get_many_to_many query handler

21057eca

Fix get_m2m_empty test

4d686c35

Replace tuple with RecordsBySubject struct

4e358f1a

Fix conflicts after rebasing on main

bb107c03

Use record_id/subject tuple as return type for get_many_to_many

90220d0b

Fix get_many_to_many pagination with composite cursor

373d0744

Fix get_many_to_many_counts pagination with fetch N+1

expand 1 comment

seoul.systems 4w

Okay. I wrapped my head around the composite cursor you proposed and am working on refactoring both storage implementations towards that. I think I might re-submit another round tomorrow :)

seoul.systems submitted #4 4w

diff interdiff

6 commits

expand

45e798b8

wip: m2m

f3144f8d

Add tests for new get_many_to_many query handler

21057eca

Fix get_m2m_empty test

4d686c35

Replace tuple with RecordsBySubject struct

4e358f1a

Fix conflicts after rebasing on main

bb107c03

Use record_id/subject tuple as return type for get_many_to_many

expand 3 comments

seoul.systems 4w

Found a bug in how we handle some of the pagination logic in cases where the number of items and the user selected limit are identical to very close too each other (already working on a fix)

bad-example.com 4w

thanks for the rebase! i tried to write things in the tiny text box but ended up needing to make a diagram: https://bsky.app/profile/did:plc:hdhoaan3xa3jiuq4fg4mefid/post/3mejuq44twc2t

key thing is that where the focus of getManyToManyCounts was the other subject (aggregation was against that, so grouping happened with it),

i think the focus of disagreggated many-to-many is on the linking records themselves

to me that takes me toward a few things

i don't think we should need to group the links by target (does the current code build up the full aggregation on every requested page? we should be able to avoid doing that)
i think the order of the response should actually be based on the linking record itself (since we have a row in the output), not the other subject, unlike with the aggregated/count version. this means you get eg. list items in order they were added instead of the order of the listed things being created. (i haven't fully wrapped my head around the grouping/ordering code here yet)
since any linking record can have a path_to_other with multiple links, i think a composite cursor could work here:

a 2-tuple of (backlink_vec_idx, forward_vec_idx).

for normal cases where the many-to-many record points to exactly one other subject, it would just be advancing backlink_vec_idx like normal backlinks

for cases where the many-to-many record actually has multiple foward links at the given path_to_other, the second part of the tuple would track progress through that list

i think that allows us to hold the necessary state between calls without needing to reconstruct too much in memory each time?

(also it's hard to write in this tiny tiny textbox and have a sense of whether what i'm saying makes sense)

seoul.systems 4w

Interesting approach! I have to think through this for a bit to be honest. Maybe I tried to follow the existing counts implementation too closely

Having said that, I added a new composite cursor to fix a couple of bugs that would arrive when hitting a couple of possible edge-cases in the pagination logic. This affects both the new get-many-to-many endpoint as well as the existing get-many-to-many-counts endpoint. As the changes are split over two distinct commits things should be straightforward to review.

Your assumption is still correct in the sense that we do indeed have to build up the aggregation again for every request. I have to double-check the get-backlinks endpoint to get a better sense of where you're going at.

Finally, I agree that the interface here doesn't necessarily make the whole thing easier to understand, unfortunately

seoul.systems submitted #3 5w

diff interdiff

6 commits

expand

dbbef261

wip: m2m

52039af4

Add tests for new get_many_to_many query handler

da6fb97c

Fix get_m2m_empty test

9bd3a217

Replace tuple with RecordsBySubject struct

6f1351b9

Fix conflicts after rebasing on main

235fb506

Use record_id/subject tuple as return type for get_many_to_many

expand 2 comments

bad-example.com 4w

i think something got funky with a rebase or the way tangled is showing it -- some of my changes on main seem to be getting shown (reverted) in the diff.

i don't mind sorting it locally but will mostly get to it tomorrow, in case you want to see what's up before i do.

seoul.systems 4w

That's one on me, sorry! Rebased again on main and now everything seems fine

seoul.systems submitted #2 7w

diff interdiff

5 commits

expand

dbbef261

wip: m2m

52039af4

Add tests for new get_many_to_many query handler

da6fb97c

Fix get_m2m_empty test

9bd3a217

Replace tuple with RecordsBySubject struct

6f1351b9

Fix conflicts after rebasing on main

expand 5 comments

seoul.systems 7w

Rebased on main. As we discussed in the PR for the order query parameter, I didn't include this here as it's not a particular sensible fit.

bad-example.com 7w

i need to get into the code properly but my initial thought is that this endpoint should return a flat list of results, like

{
  "items": [
    {
      "link": { did, collection, rkey }, // the m2m link record
      "subject": "a.com"
    },
    {
      "link": { did, collection, rkey },
      "subject": "a.com"
    },
    {
      "link": { did, collection, rkey },
      "subject": "b.com"
    },
  ]
}

this will require a bit of tricks in the cursor to track pages across half-finished groups of links

(also this isn't an immediate change request, just getting it down for discussion!)

(and separately, i've also been wondering about moving more toward returning at-uris instead of broken-out did/collection/rkey objects. which isn't specifically about this PR, but if that happens then switching before releasing it is nice)

seoul.systems 6w

Hmm, I wonder how this would then work with the path_to_other parameter. Currently we have this nested grouping in order to show and disambiguate different relationships between different links.

For instance take the following query and it's results:

http://localhost:6789/xrpc/blue.microcosm.links.getManyToMany?subject=at://did:plc:2w45zyhuklwihpdc7oj3mi63/app.bsky.feed.post/3mdbbkuq6t32y&source=app.bsky.feed.post:reply.root.uri&pathToOther=reply.parent.uri&limit=16

This query asks: "Show me all posts in this thread, grouped by who they're responding to."

A flat list would just give us all the posts in the thread. The nested structure answers a richer question: who's talking to whom? Some posts are direct responses to the original article. Others are replies to other commenters, forming side conversations that branch off from the main thread.

The pathToOther grouping preserves that distinction. Without it, we'd lose the information about who's talking to whom.

{
  "linking_records": [
    {
      "subject": "at://did:plc:2w45zyhuklwihpdc7oj3mi63/app.bsky.feed.post/3mdbbkuq6t32y",
      "records": [
        {
          "did": "did:plc:lznqwrsbnyf6fdxohikqj6h3",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd27pja7s2y"
        },
        {
          "did": "did:plc:uffx77au6hoauuuumkbuvqdr",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd2tt5efc2a"
        },
        {
          "did": "did:plc:y7qyxzo7dns5m54dlq3youu3",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd2wtjxgc2d"
        },
        {
          "did": "did:plc:yaakslxyqydb76ybgkhrr4jk",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd35hyads22"
        },
        {
          "did": "did:plc:fia7w2kbnrdjwp6zvxywt7qv",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd37j3ldk2m"
        },
        {
          "did": "did:plc:xtecipifublblkomwau5x2ok",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd3dbtbz22n"
        },
        {
          "did": "did:plc:hl5lhiy2qr4nf5e4eefldvme",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd42hpw7c2e"
        },
        {
          "did": "did:plc:fgquypfh32pewivn3bcmzseb",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd46jteoc2m"
        }
      ]
    },
    {
      "subject": "at://did:plc:3rhjxwwui6wwfokh4at3q2dl/app.bsky.feed.post/3mdczc7c4gk2i",
      "records": [
        {
          "did": "did:plc:3rhjxwwui6wwfokh4at3q2dl",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdczt7cwhk2i"
        }
      ]
    },
    {
      "subject": "at://did:plc:6buibzhkqr4vkqu75ezr2uv2/app.bsky.feed.post/3mdby25hbbk2v",
      "records": [
        {
          "did": "did:plc:fgeie2bmzlmx37iglj3xbzuj",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd26ulf4k2j"
        }
      ]
    },
    {
      "subject": "at://did:plc:lwgvv5oqh5stzb6dxa5d7z3n/app.bsky.feed.post/3mdcxqbkkfk2i",
      "records": [
        {
          "did": "did:plc:hl5lhiy2qr4nf5e4eefldvme",
          "collection": "app.bsky.feed.post",
          "rkey": "3mdd45u56sk2e"
        }
      ]
    }
  ],
  "cursor": null
}

Correct me if I'm somehow wrong here!

Regarding returning at-uris: I think this might be a nice idea as users might be able to split these up when they feel the need to any way and it feels conceptually more complete. But, it might be easier to do this in a different PR over all existing XRPC endpoints. This would allow us to add this new endpoint already while working on the updated return values in the meantime. I'd like to avoid doing too much distinct stuff in one PR. :)

bad-example.com 6w

at-uris: totally fair, holding off for a follow-up.

flat list: i might have messed it up in my example but i think what i meant is actually equivalent to the grouped version: flattened, with the subject ("group by") included with every item in the flatted list.

clients can collect the flat list and group on subject to get back to your structured example, if they want.

my motivations are probably part sql-brain, part flat-list-enjoyer, and part cursor-related. i'm trying to disregard the first two, and i'm curious about your thoughts about how to handle the cursor:

with a flat list it's easy (from the client perspective at least) -- just keep chasing the cursor for as much of the data as you need. (cursors can happen in the middle of a subject)

with nested results grouped by subject it's less obvious to me. correct me if i'm wrong (need another block of time to actually get into the code) but i think the grouped item sub-list is unbounded size in the proposed code here? so cursors are only limiting the number of groups.

if we go with the grouped nested response, i think maybe we'd want something like:

a cursor at the end for fetching more groups, and
a cursor for each group-list that lets you fetch more items from just that group-list.

(i think this kind of nested paging is pretty neat!)

seoul.systems 6w

Interesting. Now that you mention it I feel I kinda get where you're going at!

I think the whole cursor thing, albeit possible for sure, is kinda creating more unnecessary complexity so I'll probably go with your suggestion.

It seems easier to create custom groupings on their own for most users (having more freedom is always great) and I think from an ergonomic perspective the two cursors might create more friction.

seoul.systems submitted #1 2mo

diff interdiff

4 commits

expand

a4177589

wip: m2m

81b7df55

Add tests for new get_many_to_many query handler

c6742917

Fix get_m2m_empty test

f0b16b50

Replace tuple with RecordsBySubject struct

expand 1 comment

seoul.systems 2mo

Added the missing lexicon entry for the new endpoint and changed the return type as well. Commented this wrongly at the other PR that I was working on. Sorry about that lol.

seoul.systems submitted #0 2mo

diff

3 commits

expand

a4177589

wip: m2m

81b7df55

Add tests for new get_many_to_many query handler

c6742917

Fix get_m2m_empty test

expand 1 comment

seoul.systems 2mo

I think the existing get_many_to_many_counts handler and the new get_many_to_many handler are similar enough that we might extract the bulk of their logic in a shared piece of logic. Maybe a method that takes the existing identical function parameters and a new additional callback parameter (that handles what we do with found matches, i.e. calculate counts or join URIs) might be one way to go for it.

I am not too sure yet though if this is indeed the right thing to do as the new shared implementation might be a bit complicated. But given the strong similarities between the two I think it's worth at least considering.