Overview
Comment: | clean up rels mechanism, get parvan mostly working again |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | trunk |
Files: | files | file ages | folders |
SHA3-256: |
bc37f027013083243953b8bff99ce8e2 |
User & Date: | lexi on 2022-04-28 22:52:49 |
Other Links: | manifest | tags |
Context
2022-10-27
| ||
20:04 | add soda check-in: 0d4aa1c43a user: lexi tags: trunk | |
2022-04-28
| ||
22:52 | clean up rels mechanism, get parvan mostly working again check-in: bc37f02701 user: lexi tags: trunk | |
21:01 | commit to preserve old code im about to axe, parvan is broken currently check-in: f996abb5e5 user: lexi tags: trunk | |
Changes
Modified parvan.lua from [30400bd2f4] to [af37224306].
50 51 52 53 54 55 56 57 58 59 60 61 62 63 ... 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 ... 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 ... 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 ... 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 ... 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 ... 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 .... 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 .... 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 .... 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 .... 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 .... 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 .... 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 .... 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 .... 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 .... 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 .... 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 .... 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 |
if src == nil then return end local sc = #src for j=1,sc do dest[i+j] = src[j] end i = i + sc iter(...) end iter(...) end local function tcat(...) local new = {} tcatD(new, ...) return new end local ansi = { ................................................................................ decode = function(s) local n = parse(ty,s) if (n+1) > #vals then error(string.format('enum "%s" does not have %u members', table.concat(vals,'","'),n),3) end return vals[n+1] end; } end fmt.note = { {'kind', fmt.tag}; {'paras', fmt.list(fmt.string)}; } fmt.example = { {'quote',fmt.string}; {'src',fmt.label}; } fmt.meaning = { {'lit', fmt.string}; {'examples', fmt.list(fmt.example,fmt.u8)}; {'notes', fmt.list(fmt.note,fmt.u8)}; } fmt.phrase = { {'str',fmt.label}; {'means',fmt.list(fmt.meaning,fmt.u8)}; {'xref',fmt.list(fmt.path,fmt.u16)}; } fmt.def = { {'part', fmt.u8}; {'branch', fmt.list(fmt.label,fmt.u8)}; {'means', fmt.list(fmt.meaning,fmt.u8)}; {'forms', fmt.map(fmt.u16,fmt.label,fmt.u16)}; {'phrases', fmt.list(fmt.phrase,fmt.u16)}; } fmt.word = { {'defs', fmt.list(fmt.def,fmt.u8)}; } fmt.dictHeader = { {'lang', fmt.tag}; {'meta', fmt.string}; {'partsOfSpeech', fmt.list(fmt.tag,fmt.u16)}; ................................................................................ {'parts', fmt.list(fmt.tag,fmt.u8)}; -- which parts of speech does this form apply to? -- leave empty if not relevant },fmt.u16)}; } fmt.relSet = { {'uid', fmt.u32}; -- IDs are persistent random values so they can be used -- as reliable identifiers even when merging exports in -- a parvan-unaware VCS {'members', fmt.list(fmt.path,fmt.u16)}; {'kind', fmt.enum('syn','ant','met')}; } fmt.dict = { {'header', fmt.dictHeader}; {'words', fmt.map(fmt.string,fmt.word)}; {'relsets', fmt.list(fmt.relSet)}; } ................................................................................ end local function rebuildRelationCache(d) -- (re)build a dictionary's relation cache; needed -- at load time and whenever any changes to relsets -- are made (unless they're simple enough to update -- the cache directly by hand, but that's very eeeh) local sc = {} for i,s in ipairs(d.relsets) do for j,m in ipairs(s.members) do sc[m.w] = sc[m.w] or {} table.insert(sc[m.w], s) end end d._relCache = sc end local function writeDict(d) local atomizePoS, posMap = atomizer() for lit,w in pairs(d.words) do for j,def in ipairs(w.defs) do ................................................................................ if not fd then error(userError("cannot open file " .. file),2) end return fd else return file end end local function safeNavWord(ctx, word, dn, mn, nn) local w = ctx.dict.words[word] if not w then id10t 'bad word' end if dn == nil then return w end local d = w.defs[tonumber(dn)] if not d then id10t('no definition #%u',dn) end if mn == nil then return w,d end local m = d.means[tonumber(mn)] if not m then id10t('no meaning #%u',mn) end if nn == nil then return w,d,m end local n = m.notes[tonumber(nn)] if not n then id10t('no note #%u',nn) end return w,d,m,n end local function copy(tab) local new = {} for k,v in pairs(tab) do new[k] = v end return new end local function pathParse(p) ................................................................................ and a.dn == b.dn and a.mn == b.mn and a.pn == b.pn and a.nn == b.nn and a.xn == b.xn end local function pathResolve(ctx, a) if not a.w then return end -- empty paths are valid! local function lookup(seg, tbl,val) if not tbl then error('bad table',2) end local v = tbl[val] if v then return v end id10t('bad %s in path: %s', seg, val) end local res = {} res.word = lookup('word', ctx.dict.words, a.w) if not a.dn then return res end res.def = lookup('definition', w.defs, a.dn) if (not a.pn) and (not a.mn) then return res end local m if a.pn then res.phrase = lookup('phrase', d.phrases, a.pn) res.meaning = lookup('meaning', p.means, a.mn) else res.meaning = lookup('meaning', d.means, a.mn) end if a.xn then res.ex = lookup('example',m.examples,a.xn) elseif a.nn then res.note = lookup('note',m.notes,a.nn) end return res end local function pathNav(...) local t = pathResolve(...) ................................................................................ help = "add a new word"; syntax = "<word>"; write = true; exec = function(ctx,word) if ctx.dict.words[word] then id10t "word already coined" end ctx.dict.words[word] = {defs={}} end; }; def = { help = "define a word"; syntax = "<word> <part-of-speech> [<meaning> [<root>…]]"; write = true; exec = function(ctx,word,part,means,...) local etym = {...} if (not word) or not part then id10t 'bad definition' end if not ctx.dict.words[word] then ctx.dict.words[word] = {defs={}} end local n = #(ctx.dict.words[word].defs)+1 ctx.dict.words[word].defs[n] = { part = part; branch = etym; means = {means and { lit=means; examples={}; notes={}; } or nil}; forms = {}; } ctx.log('info', string.format('added definition #%u to “%s”', n, word)) end; }; mean = { help = "add a meaning to a definition"; syntax = "<word> <def#> <meaning>"; ................................................................................ end local links={} for i,l in ipairs{select(2,...)} do links[i] = pathParse(l) end local newstruct = { uid=math.random(1,0xffffFFFF); members=links; kind = rel; } table.insert(ctx.dict.relsets, newstruct) local rc = ctx.dict._relCache for i,l in pairs(links) do rc[l.w] = rc[l.w] or {} table.insert(rc[l.w], newstruct) end rebuildRelationCache(ctx.dict) else -- assemble a list of groups local tgtw = ... local wp = pathParse(tgtw) local w,d,m = pathNav(ctx, wp) for i,ss in ipairs(ctx.dict.relsets) do for j,s in ipairs(ss.members) do if pathSub(s, wp) then -- if s.word == wp.w and (wp.dn == nil or s.def == wp.dn) then table.insert(groups, {set = ss, mem = s, id = i}) break end end end if op == 'show' then for i, g in ipairs(groups) do ................................................................................ end end return repr end local others = {} for j, o in ipairs(g.set.members) do local ow = pathResolve(ctx, {w=o.w}).w if (g.set.kind == 'ant' or not pathMatch(o, g.mem)) and --exclude antonym headwords not (g.set.kind == 'ant' and j==1) then table.insert(others, ' '..label(o,ow)) end end local llab do local cdw = ctx.dict.words if g.set.kind == 'ant' then local ap = g.set.members[1] llab = fo.br(label(ap,cdw[ap.w]) or '') else llab = fo.br(label(g.mem,cdw[g.mem.w]) or '') end end local kls = { syn = fo.color('synonyms',2,true)..' of'; ant = fo.color('antonyms',1,true)..' of'; met = fo.color('metonyms',4,true)..' of'; } ................................................................................ local tgtn, paths = (select(2,...)), { select(3, ...) } rebuildRelationCache(ctx.dict) elseif op == 'destroy' then local tgtw, tgtn = ... if not tgtn then id10t 'missing group number' end local delendum = groups[tonumber(tgtn)] if not delendum then id10t 'bad group number' end local rs = ctx.dict.relsets local last = #rs if delendum.id == last then rs[delendum.id] = nil else -- since order doesn't matter, we can use a -- silly swapping trick to reduce the deletion -- worst case from O(n) to O(2) rs[delendum.id] = rs[last] rs[last] = nil end rebuildRelationCache(ctx.dict) else id10t 'invalid operation' end end end; }; ................................................................................ help = "add a note to a definition or a paragraph to a note"; syntax = {"(<m-path> (add|for) <kind> | <m-path>:<note#>) <para>…"; "m-path ::= <word>@<def#>/<meaning#>"}; write = true; exec = function(ctx,path,...) local paras, mng local dest = pathParse(path) local _,_,m = safeNavWord(ctx,dest.w,dest.dn,dest.mn) if dest.nn then paras = {...} else local op, kind = ... paras = { select(3, ...) } if op == 'add' then dest.nn = #(m.notes) + 1 m.notes[dest.nn] = {kind=kind, paras=paras} return elseif op == 'for' then for i,nn in ipairs(m.notes) do if nn.kind == kind then dest.nn = i break end end if not dest.nn then id10t('no note of kind %s in %s',kind,path) end end end local dpa = m.notes[dest.nn].paras local top = #dpa for i,p in ipairs(paras) do dpa[top+i] = p end end }; shell = { ................................................................................ end end table.sort(out, function(a,b) return a.lit < b.lit end) local fo = ctx.sty[io.stdout] local function gatherRelSets(path) local antonymSets, synonymSets, metonymSets = {},{},{} if ctx.dict._relCache[path.w] then for i, rel in ipairs(ctx.dict._relCache[path.w]) do local specuset,tgt,anto = {} for j, mbr in ipairs(rel.members) do if pathMatch(mbr, path) then if rel.kind == 'syn' then tgt = synonymSets elseif rel.kind == 'met' then tgt = metonymSets elseif rel.kind == 'ant' then if j == 1 -- is this the headword? then tgt = antonymSets else tgt = synonymSets end end elseif j == 1 and rel.kind == 'ant' then anto = mbr else table.insert(specuset, mbr) end end if tgt then table.insert(tgt, specuset) if anto then table.insert(antonymSets, {anto}) end ................................................................................ end for i, w in ipairs(out) do local d = fo.ul(fo.br(w.lit)) local wordrels = autobreak(table.concat( formatRels(gatherRelSets{w=w.lit}, 2), '\n' )) local wc = ctx.dict._relCache[w.lit] if #w.word.defs == 1 then d=d .. ' ' .. fo.rgb(fo.em('('..(w.word.defs[1].part)..')'),.8,.5,1) .. '\n' .. meanings(w,w.word.defs[1],false,1) .. '\n' .. autobreak(table.concat(formatRels(gatherRelSets{w=w.lit,dn=1}, 4), '\n')) .. wordrels .. '\n' else ................................................................................ if c ~= 'pv0' then id10t "not a parvan export" end new.header.lang = words[2] new.header.meta = words[3] state = 1 else print(pathString(path, ctx.sty[io.stderr])) local W,D,M,N = pathNav({dict=new}, path) if c == 'w' then syn(1) state = 2 path = {w=words[2]} new.words[words[2]] = {defs={}} elseif c == 'f' then syn(1) local nf = { name = words[2]; abbrev = words[3] or ""; desc = words[4] or ""; parts = {}; } ................................................................................ table.insert(lastinfl.parts,words[2]) elseif c == 's' then syn(2) relsets[words[3]] = relsets[words[3]] or {} relsets[words[3]].kind = words[2] relsets[words[3]].uid = tonumber(words[3]) relsets[words[3]].members = relsets[words[3]].members or {} elseif state >= 2 and c == 'r' then syn(1) relsets[words[2]] = relsets[words[2]] or { uid = tonumber(words[2]); members={}; } table.insert(relsets[words[2]].members, path) elseif state >= 2 and c == 'd' then syn(1) state = 3 table.insert(W.defs, { part = words[2]; branch = {}; means = {}; forms = {}; phrases = {}; }) path = {w = path.w, dn = #(W.defs)} elseif state >= 3 and c == 'dr' then syn(1) table.insert(D.branch, words[2]) elseif state >= 3 and c == 'df' then syn(2) if not inflmap[words[2]] then id10t('no inflection form %s defined', words[2]) end D.forms[inflmap[words[2]]] = words[3] elseif state >= 3 and c == 'm' then syn(1) state = 4 table.insert(D.means, { lit = words[2]; notes = {}; examples = {}; }); path = {w = path.w, dn = path.dn, mn = #(D.means)} elseif state >= 4 and c == 'n' then syn(1) state = 5 table.insert(M.notes, {kind=words[2], paras={}}) path = {w = path.w, dn = path.dn, mn = path.mn, nn = #(M.notes)}; elseif state >= 5 and c == 'np' then syn(1) table.insert(N.paras, words[2]) end -- we ignore invalid ctls, for sake of forward-compat end end end ................................................................................ for k,v in pairs(relsets) do if not v.uid then --handle non-numeric export ids v.uid = math.random(0,0xffffFFFF) end table.insert(new.relsets, v) end local ofd = safeopen(ctx.file,"w+b") local o = writeDict(new); ofd:write(o) ofd:close() end ................................................................................ if ctx.flags.human and lvl > 0 then pfx = string.rep('\t', lvl) end ofd:write(pfx..string.format(...)..'\n') end local d = ctx.dict o(0,'pv0 %s %s', san(d.header.lang), san(d.header.meta)) local function checksyn(obj) -- for _,s in ipairs(d.synonyms) do local lvl = 0 if obj.nn then lvl = 4 elseif obj.mn then lvl = 3 elseif obj.dn then lvl = 2 elseif obj.w then lvl = 1 end if not d._relCache[obj.w] then return end for _,s in ipairs(d._relCache[obj.w]) do for _,sm in ipairs(s.members) do if pathMatch(obj, sm) then o(lvl,'r %u',s.uid) break end end end end for i,f in pairs(d.header.inflectionForms) do o(0,'f %s %s %s', san(f.name), san(f.abbrev), san(f.desc)) for j,p in pairs(f.parts) do o(1,'fp %s', san(p)) end end for lit, w in pairs(d.words) do o(0,'w %s',san(lit)) checksyn{w=lit} for i,def in ipairs(w.defs) do o(1,'d %s',san(def.part)) checksyn{w=lit,dn=i} for j,r in ipairs(def.branch) do o(2,'dr %s',san(r)) end for j,m in ipairs(def.means) do o(2,'m %s', san(m.lit)) checksyn{w=lit,dn=i,mn=j} for k,n in ipairs(m.notes) do o(3,'n %s', san(n.kind)) for a,p in ipairs(n.paras) do o(4,'np %s', san(p)) end end end end end for _,s in ipairs(d.relsets) do o(0,'s %s %u', s.kind, s.uid) end end local function filterD(lst, fn) -- cheap algorithm to destructively filter a list ................................................................................ top = top - 1 end end return lst end function cmds.mod.exec(ctx, orig, oper, dest, ...) if (not orig) or not oper then id10t '`mod` requires at least an origin and an operation' end local op, dp = pathParse(orig) local w,d,m,n = safeNavWord(ctx, op.w,op.dn,op.mn,op.nn) -- unfortunately, "pointers" exist elsewhere in the -- structure, currently just from relsets, that must -- be updated whenever an object moves or changes. -- this is miserable and takes a lot of work, using -- algorithms provided by the following functions. -- note that we don't bother trying to update the -- relcache as we go, it's simply not worth the work; -- instead we simply rebuild the whole cache when -- this command returns local function cleanupRels(path, fn) local rc = ctx.dict._relCache[path.w] if rc then for k,s in pairs(rc) do fn(s,k) end end end local function cleanupRelsEach(path, fn) cleanupRels(path, function(s,k) local top = #s.members for i=1,top do local m=s.members[i] if pathSub(path, m) then local val = fn(m,s,i) if val ~= nil then s.members[i] = val end end end end) end local function deleteRefsTo(path) cleanupRels(path, function(s) -- antonyms: delete the headword and transform the group -- to a list of synonyms if s.kind == 'ant' and pathSub(path,s.members[1]) then s.kind = 'syn' end filterD(s.members, function(m) return not pathSub(path, m) end) end) if not path.dn then ctx.dict._relCache[path.w] = nil end end local function moveRelTree(op,dp) cleanupRelsEach(op, function(old,set,idx) local new = {} for _,elt in pairs{'w','dn','mn','nn'} do if dp[elt] ~= nil then new[elt] = dp[elt] else new[elt] = op[elt] or old[elt] end end return new end) end local function shiftRelTree(dp, fld, nid, amt) local cleanupTargetMask = ({ dn = {w=dp.w}; mn = {w=dp.w,dn=dp.dn}; nn = {w=dp.w,dn=dp.dn,mn=dp.mn}; })[fld] -- >____< cleanupRelsEach(cleanupTargetMask, function(old,set,i) if old[fld] >= nid then old[fld] = old[fld] + amt end end) end local function insertAndMoveRelTree(tbl,n,op,dp,fld) local nid = #tbl local path = copy(dp) path[fld] = nid tbl[nid] = n shiftRelTree(dp,fld,1) moveRelTree(op, path) end if oper == 'drop' then -- clean out the cache and delete relationships deleteRefsTo(op) if not d then ctx.dict.words[op.w] = nil elseif not m then table.remove(w.defs, op.dn) elseif not n then table.remove(d.means, op.mn) else table.remove(m.notes, op.nn) end elseif oper == 'out' then if n or not m then id10t '`mod out` must target a meaning' end if not dest then id10t '`mod out` requires at least a part of speech' end local newdef = { part = dest; branch = {...}; forms = {}; means = {m}; } shiftRelTree(op, 'dn', op.dn, 1) table.insert(w.defs,op.dn+1, newdef) moveRelTree(op,{w=op.w, dn=op.dn+1, mn=1}) table.remove(d.means,op.mn) elseif oper == 'move' or oper == 'merge' or oper == 'clobber' then if dest then dp = pathParse(dest) else id10t('`mod %s` requires a target',oper) end if n then if not dp.mn then id10t '`mod` on a note requires a note or meaning destination' end local _,_,dm = safeNavWord(ctx, dp.w,dp.dn,dp.mn) if dp.nn then if oper == 'move' then shiftRelTree(dp, 'nn', dp.nn, 1) table.insert(dm.notes, dp.nn, n) elseif oper == 'merge' then local top = #(dm.notes[dp.nn].paras) for i, v in ipairs(n.paras) do dm.notes[dp.nn].paras[i+top] = v end elseif oper == 'clobber' then deleteRefsTo(dp) dm.notes[dp.nn] = n end moveRelTree(op,dp) else if oper ~= 'move' then id10t('`mod note %s` requires a note target', oper) end insertAndMoveRelTree(dm.notes,n,op,dp,'nn') end if oper == 'move' and dp.nn and dm == m and op.nn > dp.nn then table.remove(m.notes,op.nn+1) else table.remove(m.notes,op.nn) end elseif m then if not dp.dn then local newdef = { part = d.part; branch = copy(d.branch); forms = copy(d.forms); means = {m}; } local didx if ctx.dict.words[dp.w] then local defst = ctx.dict.words[dp.w].defs didx = #defst defst[didx] = newdef else ctx.dict.words[dp.w] = { defs = {newdef}; } didx = 1 end cleanupRelsEach(op, function(oldpath,set,mi) return {w=dp.w,dn=didx,mn=1,nn=oldpath.nn} end) table.remove(d.means,dp.mn) else local dw, dd = safeNavWord(ctx, dp.w, dp.dn) if dp.mn then if dd.means[dp.mn] and (oper == 'merge' or oper=='clobber') then if oper == 'merge' then dd.means[dp.mn] = dd.means[dp.mn] .. '; ' .. m elseif oper == 'clobber' then deleteRefsTo(dp) dd.means[dp.mn] = m end else cleanupRelsEach({w=dp.w,dn=dp.dn}, function(old,set,i) if old.mn >= dp.mn then old.mn = old.mn + 1 end end) table.insert(dd.means, dp.mn, m) end moveRelTree(op,dp) else insertAndMoveRelTree(dd.means,m, op,dp,'mn') -- table.insert(dd.means, m) end if oper == 'move' and dp.mn and dd.means == d.means and op.mn > dp.mn then table.remove(d.means,op.mn+1) else table.remove(d.means,op.mn) end end elseif d then local ddefs = safeNavWord(ctx, dp.w).defs if dp.dn then if oper == 'merge' then local top = #(ddefs[dp.dn].means) for i,om in ipairs(d.means) do ddefs[dp.dn].means[top+i] = om end for k,p in pairs(d.forms) do deleteRefsTo(dp) ddefs[dp.dn].forms[k] = p -- clobbers! end else shiftRelTree(dp,'dn',dp.dn,1) table.insert(ddefs, dp.dn, d) end moveRelTree(op,dp) else insertAndMoveRelTree(ddefs,d, op,dp,'dn') -- table.insert(ddefs, d) end if oper == 'move' and dp.mn and w.defs == ddefs and op.mn > dp.mn then table.remove(w.defs,op.dn+1) else table.remove(w.defs,op.dn) end else if ctx.dict.words[dp.w] then if oper ~= 'merge' then id10t('the word “%s” already exists; use `merge` if you want to merge the words together', dp.w) end for i,def in ipairs(w.defs) do local odp = copy(op) odp.dn = i local ddp = {w=dp.w, dn=dp.dn+i-1} if dp.dn then shiftRelTree(dp, 'dn', dp.dn+i-1, 1) table.insert(ctx.dict.words[dp.w].defs, dp.dn+i-1, def) moveRelTree(odp,ddp) else -- table.insert(ctx.dict.words[dp.w].defs, def) insertAndMoveRelTree(ctx.dict.words[dp.w].defs, def, odp,dp,'dn') end end else ctx.dict.words[dp.w] = w moveRelTree(op,dp) -- ctx.dict._relCache[dp.w] = ctx.dict._relCache[op.w] -- ctx.dict._relCache[op.w] = nil end ctx.dict.words[op.w] = nil end end rebuildRelationCache(ctx.dict) end local function fileLegible(file) -- check if we can access the file local fd = io.open(file,"rb") local ret = false |
> > > > > > > > > > > > > > > > | | < | | | < > > | | | | | > > > > > > > > > > > > > > > > > > > | < < < < < < < < < < < < < < < < < < < > > | < | | | | | | | | > > > < | | | | | < | > | > | | < | > > > > > > > > | > | | | | | | | | | | | | | | | | | | | | > > | | | | < < | > > > > > > > | > > > > > > > > > > | > > > > > > > > > | | > > > > | | < < < < < < < < < | < < < > > > > > > > > > > > > > > | | | | | < < < < | < < > < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < |
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 ... 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 ... 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 ... 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 ... 784 785 786 787 788 789 790 791 792 793 794 795 796 797 ... 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 ... 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 .... 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 .... 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 .... 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 .... 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 .... 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 .... 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 .... 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 .... 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 .... 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 .... 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 .... 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 |
if src == nil then return end local sc = #src for j=1,sc do dest[i+j] = src[j] end i = i + sc iter(...) end iter(...) end local function fastDelete(table,idx) -- delete without preserving table order local l = #table table[idx] = table[l] table[l] = nil return table end local function tcat(...) local new = {} tcatD(new, ...) return new end local ansi = { ................................................................................ decode = function(s) local n = parse(ty,s) if (n+1) > #vals then error(string.format('enum "%s" does not have %u members', table.concat(vals,'","'),n),3) end return vals[n+1] end; } end fmt.uid = fmt.u32 fmt.relatable = function(ty) return tcat(ty,{ {'rels',fmt.list(fmt.uid,fmt.u16)}; }) end fmt.note = { {'kind', fmt.tag}; {'paras', fmt.list(fmt.string)}; } fmt.example = { {'quote',fmt.string}; {'src',fmt.label}; } fmt.meaning = fmt.relatable { {'lit', fmt.string}; {'examples', fmt.list(fmt.example,fmt.u8)}; {'notes', fmt.list(fmt.note,fmt.u8)}; } fmt.phrase = fmt.relatable { {'str',fmt.label}; {'means',fmt.list(fmt.meaning,fmt.u8)}; } fmt.def = fmt.relatable { {'part', fmt.u8}; {'branch', fmt.list(fmt.label,fmt.u8)}; {'means', fmt.list(fmt.meaning,fmt.u8)}; {'forms', fmt.map(fmt.u16,fmt.label,fmt.u16)}; {'phrases', fmt.list(fmt.phrase,fmt.u16)}; } fmt.word = fmt.relatable { {'defs', fmt.list(fmt.def,fmt.u8)}; } fmt.dictHeader = { {'lang', fmt.tag}; {'meta', fmt.string}; {'partsOfSpeech', fmt.list(fmt.tag,fmt.u16)}; ................................................................................ {'parts', fmt.list(fmt.tag,fmt.u8)}; -- which parts of speech does this form apply to? -- leave empty if not relevant },fmt.u16)}; } fmt.relSet = { {'uid', fmt.uid}; -- IDs are persistent random values so they can be used -- as reliable identifiers even when merging exports in -- a parvan-unaware VCS {'kind', fmt.enum('syn','ant','met')}; -- membership is stored in individual objects, using a field -- attached by the 'relatable' template } fmt.dict = { {'header', fmt.dictHeader}; {'words', fmt.map(fmt.string,fmt.word)}; {'relsets', fmt.list(fmt.relSet)}; } ................................................................................ end local function rebuildRelationCache(d) -- (re)build a dictionary's relation cache; needed -- at load time and whenever any changes to relsets -- are made (unless they're simple enough to update -- the cache directly by hand, but that's very eeeh) local setMems = {} -- indexed by set id local function scan(obj,path) for _,v in pairs(obj.rels) do setMems[v] = setMems[v] or {mems={}} table.insert(setMems[v].mems, {path=path, obj=obj}) end end for wk,wv in pairs(d.words) do scan(wv, {w=wk}) for dk,dv in pairs(wv.defs) do scan(dv, {w=wk, dn=dk}) for mk,mv in pairs(dv.means) do scan(mv, {w=wk, dn=dk, mn=mk}) end for pk,pv in pairs(dv.phrases) do scan(pv, {w=wk, dn=dk, pn=pk}) for mk,mv in pairs(pv.means) do scan(mv, {w=wk, dn=dk, pn=pk, mn=mk}) end end end end for sk,sv in pairs(d.relsets) do setMems[sv.uid] = setMems[sv.uid] or {} setMems[sv.uid].set = sv end d._relCache = setMems end local function writeDict(d) local atomizePoS, posMap = atomizer() for lit,w in pairs(d.words) do for j,def in ipairs(w.defs) do ................................................................................ if not fd then error(userError("cannot open file " .. file),2) end return fd else return file end end local function copy(tab) local new = {} for k,v in pairs(tab) do new[k] = v end return new end local function pathParse(p) ................................................................................ and a.dn == b.dn and a.mn == b.mn and a.pn == b.pn and a.nn == b.nn and a.xn == b.xn end local function pathResolve(ctx, a) local res = {} if not a.w then return res end -- empty paths are valid! local function lookup(seg, tbl,val) if not tbl then error('bad table',2) end local v = tbl[val] if v then return v end id10t('bad %s in path: %s', seg, val) end res.word = lookup('word', ctx.dict.words, a.w) if not a.dn then return res end res.def = lookup('definition', res.word.defs, a.dn) if (not a.pn) and (not a.mn) then return res end local m if a.pn then res.phrase = lookup('phrase', res.def.phrases, a.pn) res.meaning = lookup('meaning', res.phrase.means, a.mn) else res.meaning = lookup('meaning', res.def.means, a.mn) end if a.xn then res.ex = lookup('example',res.meaning.examples,a.xn) elseif a.nn then res.note = lookup('note',res.meaning.notes,a.nn) end return res end local function pathNav(...) local t = pathResolve(...) ................................................................................ help = "add a new word"; syntax = "<word>"; write = true; exec = function(ctx,word) if ctx.dict.words[word] then id10t "word already coined" end ctx.dict.words[word] = {defs={},rels={}} end; }; def = { help = "define a word"; syntax = "<word> <part-of-speech> [<meaning> [<root>…]]"; write = true; exec = function(ctx,word,part,means,...) local etym = {...} if (not word) or not part then id10t 'bad definition' end if not ctx.dict.words[word] then ctx.dict.words[word] = {defs={},rels={}} end local n = #(ctx.dict.words[word].defs)+1 ctx.dict.words[word].defs[n] = { part = part; branch = etym; means = {means and { lit=means; examples={}; notes={}; rels={}; } or nil}; forms = {}; phrases = {}; rels={}; } ctx.log('info', string.format('added definition #%u to “%s”', n, word)) end; }; mean = { help = "add a meaning to a definition"; syntax = "<word> <def#> <meaning>"; ................................................................................ end local links={} for i,l in ipairs{select(2,...)} do links[i] = pathParse(l) end local newstruct = { uid=math.random(1,0xffffFFFF); kind = rel; } table.insert(ctx.dict.relsets, newstruct) for _, m in pairs(links) do local obj = pathRef(ctx,m) table.insert(obj.rels,newstruct.uid) end rebuildRelationCache(ctx.dict) else -- assemble a list of groups local tgtw = ... local wp = pathParse(tgtw) local o = pathResolve(ctx, wp) for i,rs in pairs(ctx.dict.relsets) do local allMembers = ctx.dict._relCache[rs.uid].mems for j,s in ipairs(allMembers) do if pathSub(s.path, wp) then table.insert(groups, { set = { uid = rs.uid; kind = rs.kind; members = allMembers; }; mem = s; id = i; }) break end end end if op == 'show' then for i, g in ipairs(groups) do ................................................................................ end end return repr end local others = {} for j, oo in ipairs(g.set.members) do local o = oo.path local ow = pathResolve(ctx, {w=o.w}).w if (g.set.kind == 'ant' or not pathMatch(o, g.mem.path)) and --exclude antonym headwords not (g.set.kind == 'ant' and j==1) then table.insert(others, ' '..label(o,ow)) end end local llab do local cdw = ctx.dict.words if g.set.kind == 'ant' then local ap = g.set.members[1].path llab = fo.br(label(ap,cdw[ap.w]) or '') else llab = fo.br(label(g.mem.path,cdw[g.mem.w]) or '') end end local kls = { syn = fo.color('synonyms',2,true)..' of'; ant = fo.color('antonyms',1,true)..' of'; met = fo.color('metonyms',4,true)..' of'; } ................................................................................ local tgtn, paths = (select(2,...)), { select(3, ...) } rebuildRelationCache(ctx.dict) elseif op == 'destroy' then local tgtw, tgtn = ... if not tgtn then id10t 'missing group number' end local delendum = groups[tonumber(tgtn)] if not delendum then id10t 'bad group number' end for k,v in pairs(delendum.set.members) do for idx, e in pairs(v.obj.rels) do if e == delendum.set.uid then fastDelete(v.obj.rels,idx) end end end fastDelete(ctx.dict.relsets, delendum.id) rebuildRelationCache(ctx.dict) else id10t 'invalid operation' end end end; }; ................................................................................ help = "add a note to a definition or a paragraph to a note"; syntax = {"(<m-path> (add|for) <kind> | <m-path>:<note#>) <para>…"; "m-path ::= <word>@<def#>/<meaning#>"}; write = true; exec = function(ctx,path,...) local paras, mng local dest = pathParse(path) local t = pathResolve(ctx,path) if dest.nn then paras = {...} else local op, kind = ... paras = { select(3, ...) } if op == 'add' then dest.nn = #(t.m.notes) + 1 t.m.notes[dest.nn] = {kind=kind, paras=paras} return elseif op == 'for' then for i,nn in ipairs(t.m.notes) do if nn.kind == kind then dest.nn = i break end end if not dest.nn then id10t('no note of kind %s in %s',kind,path) end end end local dpa = t.m.notes[dest.nn].paras local top = #dpa for i,p in ipairs(paras) do dpa[top+i] = p end end }; shell = { ................................................................................ end end table.sort(out, function(a,b) return a.lit < b.lit end) local fo = ctx.sty[io.stdout] local function gatherRelSets(path) local antonymSets, synonymSets, metonymSets = {},{},{} local obj = pathRef(ctx,path) if next(obj.rels) then for i, relid in ipairs(obj.rels) do local specuset,tgt,anto = {} local rel = ctx.dict._relCache[relid].set for j, mbr in ipairs(ctx.dict._relCache[relid].mems) do if pathMatch(mbr.path, path) then if rel.kind == 'syn' then tgt = synonymSets elseif rel.kind == 'met' then tgt = metonymSets elseif rel.kind == 'ant' then if j == 1 -- is this the headword? then tgt = antonymSets else tgt = synonymSets end end elseif j == 1 and rel.kind == 'ant' then anto = mbr.path else table.insert(specuset, mbr.path) end end if tgt then table.insert(tgt, specuset) if anto then table.insert(antonymSets, {anto}) end ................................................................................ end for i, w in ipairs(out) do local d = fo.ul(fo.br(w.lit)) local wordrels = autobreak(table.concat( formatRels(gatherRelSets{w=w.lit}, 2), '\n' )) if #w.word.defs == 1 then d=d .. ' ' .. fo.rgb(fo.em('('..(w.word.defs[1].part)..')'),.8,.5,1) .. '\n' .. meanings(w,w.word.defs[1],false,1) .. '\n' .. autobreak(table.concat(formatRels(gatherRelSets{w=w.lit,dn=1}, 4), '\n')) .. wordrels .. '\n' else ................................................................................ if c ~= 'pv0' then id10t "not a parvan export" end new.header.lang = words[2] new.header.meta = words[3] state = 1 else local T = pathResolve({dict=new}, path) local W,D,P,M,N,X = T.word, T.def, T.phrase, T.meaning, T.note, T.ex if c == 'w' then syn(1) state = 2 path = {w=words[2]} new.words[words[2]] = {defs={},rels={}} elseif c == 'f' then syn(1) local nf = { name = words[2]; abbrev = words[3] or ""; desc = words[4] or ""; parts = {}; } ................................................................................ table.insert(lastinfl.parts,words[2]) elseif c == 's' then syn(2) relsets[words[3]] = relsets[words[3]] or {} relsets[words[3]].kind = words[2] relsets[words[3]].uid = tonumber(words[3]) relsets[words[3]].members = relsets[words[3]].members or {} elseif state >= 2 and c == 'r' then syn(1) local rt if state == 2 then rt = W.rels elseif state == 3 then rt = D.rels elseif state == 4 then rt = D.rels elseif state == 14 then rt = P.rels end relsets[words[2]] = relsets[words[2]] or { uid = tonumber(words[2]) or math.random(0,0xffffFFFF); members={}; } table.insert(relsets[words[2]].members, path) elseif state >= 2 and c == 'd' then syn(1) state = 3 table.insert(W.defs, { part = words[2]; branch = {}; means = {}; forms = {}; phrases = {}; rels = {}; }) path = {w = path.w, dn = #(W.defs)} elseif state >= 3 and c == 'dr' then syn(1) table.insert(D.branch, words[2]) elseif state >= 3 and c == 'df' then syn(2) if not inflmap[words[2]] then id10t('no inflection form %s defined', words[2]) end D.forms[inflmap[words[2]]] = words[3] elseif state >= 3 and c == 'p' then syn(1) state = 14 table.insert(D.phrases, { str = words[2]; means = {}; rels = {}; }) path = {w = path.w, dn = path.dn, pn = #(D.phrases)} elseif state >= 3 and c == 'm' then syn(1) state = 4 table.insert(D.means, { lit = words[2]; notes = {}; examples = {}; rels = {}; }); path = {w = path.w, dn = path.dn, pn=path.pn, mn = #(D.means)} elseif state >= 4 and c == 'n' then syn(1) state = 5 table.insert(M.notes, {kind=words[2], paras={}}) path = {w = path.w, dn = path.dn, pn = path.pn, mn = path.mn, nn = #(M.notes)}; elseif state >= 5 and c == 'np' then syn(1) table.insert(N.paras, words[2]) end -- we ignore invalid ctls, for sake of forward-compat end end end ................................................................................ for k,v in pairs(relsets) do if not v.uid then --handle non-numeric export ids v.uid = math.random(0,0xffffFFFF) end table.insert(new.relsets, v) for q,m in pairs(v.members) do table.insert(pathRef({dict=new},m).rels, v.uid) end end local ofd = safeopen(ctx.file,"w+b") local o = writeDict(new); ofd:write(o) ofd:close() end ................................................................................ if ctx.flags.human and lvl > 0 then pfx = string.rep('\t', lvl) end ofd:write(pfx..string.format(...)..'\n') end local d = ctx.dict o(0,'pv0 %s %s', san(d.header.lang), san(d.header.meta)) local function checksyn(obj,lvl) for k,v in pairs(obj.rels) do o(lvl,'r %u',s.uid) end end for i,f in pairs(d.header.inflectionForms) do o(0,'f %s %s %s', san(f.name), san(f.abbrev), san(f.desc)) for j,p in pairs(f.parts) do o(1,'fp %s', san(p)) end end local function scanMeans(tbl,path,lvl) for j,m in ipairs(def.means) do o(lvl,'m %s', san(m.lit)) local lp = copy(path) lp.mn = j checksyn(m,lp,lvl+1) for k,n in ipairs(m.notes) do o(lvl+1,'n %s', san(n.kind)) for a,p in ipairs(n.paras) do o(lvl+2,'np %s', san(p)) end end end end for lit, w in pairs(d.words) do o(0,'w %s',san(lit)) checksyn(w,{w=lit},1) for i,def in ipairs(w.defs) do o(1,'d %s',san(def.part)) checksyn(def,{w=lit,dn=i},2) for j,r in ipairs(def.branch) do o(2,'dr %s',san(r)) end for j,p in ipairs(def.phrases) do o(2,'p %s',san(p.str)) scanMeans(p.means, {w=lit,dn=i,pn=j}, 3) end scanMeans(def.means, {w=lit,dn=i}, 2) end end for _,s in ipairs(d.relsets) do o(0,'s %s %u', s.kind, s.uid) end end local function filterD(lst, fn) -- cheap algorithm to destructively filter a list ................................................................................ top = top - 1 end end return lst end function cmds.mod.exec(ctx, orig, oper, dest, ...) rebuildRelationCache(ctx.dict) end local function fileLegible(file) -- check if we can access the file local fd = io.open(file,"rb") local ret = false |