@graphty/graph-io 0.3.40 → 0.3.41

package/dist/csv.js

@@ -1,8 +1,8 @@
-import { b as plural, bU as DICT_SAMPLE_ROWS, ac as DictHeuristic, R as ROLE_TAKEN_CODE$1, I as ID_MERGED_CODE$1, m as DUPLICATE_EDGE_ID_CODE$1, n as DUPLICATE_NODE_CODE$1, t as MISSING_ID_CODE$1, s as MISSING_ENDPOINT_CODE$1, E as EMPTY_INPUT_CODE$1, al as resolveImportOptions, a8 as ImportReportBuilder, ao as reportSinkOptions, an as reportUnusedOptions, aq as DirectionResolver, ar as IdCoercer, am as throwIfAborted, O as OPTION_IGNORED_CODE, aK as parseWeightText, a0 as uniqueColumnName, ay as RENAMED_CODE, a1 as declareResolved, c as capabilities, L as LOSS, a as checkCapabilities, r as resolveExportOptions, p as pairFolding, bD as countMixedEdges, $ as agree, e as explicitWeights, b_ as otherFormatDialect, bS as joinListText, aG as formatInteger, aF as formatDecimal, aH as canonicalId, M as MIXED_DIRECTION_CODE, D as DIRECTION_FORCED_CODE, d as DIRECTION_REFUSED_CODE, S as SINK_OPTION_CODE, C as COLUMN_RENAMED_CODE, w as UNKNOWN_ENCODING_CODE, x as ENCODING_FALLBACK_CODE, y as INVALID_ENCODING_CODE, z as INVALID_UTF8_CODE, P as PRECISION_CODE, F as ELEMENT_ISSUE, G as INPUT_ISSUE } from "./chunks/export-B-aVCwyG.js";
-import { i as inferTextDtype, T as TextCellWriter, p as parseTextCell, W as WIDENING_UNSUPPORTED_CODE } from "./chunks/text-Bjfa1xYL.js";
+import { b as plural, bU as DICT_SAMPLE_ROWS, ac as DictHeuristic, R as ROLE_TAKEN_CODE$1, I as ID_MERGED_CODE$1, m as DUPLICATE_EDGE_ID_CODE$1, n as DUPLICATE_NODE_CODE$1, t as MISSING_ID_CODE$1, s as MISSING_ENDPOINT_CODE$1, E as EMPTY_INPUT_CODE$1, al as resolveImportOptions, a8 as ImportReportBuilder, ao as reportSinkOptions, an as reportUnusedOptions, aq as DirectionResolver, ar as IdCoercer, am as throwIfAborted, O as OPTION_IGNORED_CODE, aK as parseWeightText, a0 as uniqueColumnName, ay as RENAMED_CODE, a1 as declareResolved, c as capabilities, L as LOSS, a as checkCapabilities, r as resolveExportOptions, p as pairFolding, bD as countMixedEdges, $ as agree, e as explicitWeights, b_ as otherFormatDialect, bS as joinListText, aG as formatInteger, aF as formatDecimal, aH as canonicalId, M as MIXED_DIRECTION_CODE, D as DIRECTION_FORCED_CODE, d as DIRECTION_REFUSED_CODE, S as SINK_OPTION_CODE, C as COLUMN_RENAMED_CODE, w as UNKNOWN_ENCODING_CODE, x as ENCODING_FALLBACK_CODE, y as INVALID_ENCODING_CODE, z as INVALID_UTF8_CODE, P as PRECISION_CODE, F as ELEMENT_ISSUE, G as INPUT_ISSUE } from "./chunks/export-DD3Lvw-F.js";
+import { i as inferTextDtype, T as TextCellWriter, p as parseTextCell, W as WIDENING_UNSUPPORTED_CODE } from "./chunks/text-Bh9SaAob.js";
 import { GraphFormatError, INVALID_INDEX } from "@graphty/graph-format";
-import { s as stripLeadingComments, a as sniffNewline, b as sniffDelimiter, D as DELIMITER_CANDIDATES, c as checkRecordSyntax, C as CsvRecordReader, d as splitRecords, B as BAD_QUOTE_CODE, U as UNCLOSED_QUOTE_CODE } from "./chunks/records-r3SFHz11.js";
-import { o as quoteCsvCell } from "./chunks/escape-BJRLd3vJ.js";
+import { s as stripLeadingComments, a as sniffNewline, b as sniffDelimiter, D as DELIMITER_CANDIDATES, c as checkRecordSyntax, C as CsvRecordReader, d as splitRecords, B as BAD_QUOTE_CODE, U as UNCLOSED_QUOTE_CODE } from "./chunks/records-Uc3LkABx.js";
+import { o as quoteCsvCell } from "./chunks/escape-DBb8zuNE.js";
 import { j as joinText, e as encodeChunks } from "./chunks/writer-DxjgHwSe.js";
 const SOURCE_NAMES = Object.freeze([
   "source",
@@ -433,6 +433,15 @@ function isBlank(text) {
 function isUnset(text, quoted) {
   return quoted !== true && isBlank(text);
 }
+function setNeighborCells(row, quoted) {
+  let neighbors = 0;
+  for (let k = 1; k < row.length; k++) {
+    if (!isUnset(row[k], quoted[k])) {
+      neighbors++;
+    }
+  }
+  return neighbors;
+}
 const ABORT_CHECK_INTERVAL = 64;
 function uniqueNames(names, report, line) {
   const seen = /* @__PURE__ */ new Set();
@@ -1199,12 +1208,7 @@ class TableReader {
     const { where } = this;
     where.line = line;
     where.element = null;
-    let neighbors = 0;
-    for (let k = 1; k < row.length; k++) {
-      if (!isUnset(row[k], quoted[k])) {
-        neighbors++;
-      }
-    }
+    const neighbors = setNeighborCells(row, quoted);
     if (isUnset(row[0], quoted[0])) {
       report.error("missing-value", MISSING_ID_CODE, `blank node cell`, { line });
       counts.skippedNodes++;
@@ -1926,19 +1930,23 @@ function nodeTableNote(names, table, notes) {
     );
   }
 }
+function idTextCollisions(ids) {
+  const seen = /* @__PURE__ */ new Set();
+  let collisions = 0;
+  for (let i = 0; i < ids.size; i++) {
+    const text = String(ids.idOf(i));
+    if (seen.has(text)) {
+      collisions++;
+    } else {
+      seen.add(text);
+    }
+  }
+  return collisions;
+}
 function idNotes(snapshot, values, note) {
   const { ids } = snapshot;
   if (ids.kind === "mixed") {
-    const seen = /* @__PURE__ */ new Set();
-    let collisions = 0;
-    for (let i = 0; i < ids.size; i++) {
-      const text = String(ids.idOf(i));
-      if (seen.has(text)) {
-        collisions++;
-      } else {
-        seen.add(text);
-      }
-    }
+    const collisions = idTextCollisions(ids);
     if (collisions > 0) {
       note(
         CSV_LOSS.ID_TEXT_COLLISION,
@@ -2152,39 +2160,44 @@ function attributeColumns(table, domain, reserved, idColumn, labelColumn, note)
     if (role !== null && SKIPPED_ROLES.has(role)) {
       continue;
     }
-    const low = lower(name);
     const set = column.length - column.nullCount;
-    const isIdName = domain === "edge" ? findColumn([name], EDGE_ID_NAMES) >= 0 : false;
-    if (reserved.has(low) || isIdName || idColumn !== null && lower(idColumn.meta.name) === low) {
-      note?.(
-        CSV_LOSS.RESERVED_NAME,
-        `${domain} column "${name}" is not written: the name is reserved for the ${isIdName ? "edge id" : low} column`,
-        name,
-        set
-      );
+    if (isReservedName(name, domain, reserved, idColumn, set, note)) {
       continue;
     }
-    if (findColumn([name], LABEL_NAMES) >= 0) {
-      if (labelColumn !== null) {
-        note?.(
-          CSV_LOSS.RESERVED_NAME,
-          `${domain} column "${name}" is not written: the name is reserved for the label column`,
-          name,
-          set
-        );
-        continue;
-      }
-      note?.(
-        CSV_LOSS.ROLE_ASSUMED,
-        `${domain} column "${name}" reads back with the label role (string)`,
-        name,
-        set
-      );
+    if (findColumn([name], LABEL_NAMES) >= 0 && !labelNameWritten(name, domain, labelColumn, set, note)) {
+      continue;
     }
     out.push({ column, header: name });
   }
   return out;
 }
+function isReservedName(name, domain, reserved, idColumn, set, note) {
+  const low = lower(name);
+  const isIdName = domain === "edge" ? findColumn([name], EDGE_ID_NAMES) >= 0 : false;
+  if (reserved.has(low) || isIdName || idColumn !== null && lower(idColumn.meta.name) === low) {
+    note?.(
+      CSV_LOSS.RESERVED_NAME,
+      `${domain} column "${name}" is not written: the name is reserved for the ${isIdName ? "edge id" : low} column`,
+      name,
+      set
+    );
+    return true;
+  }
+  return false;
+}
+function labelNameWritten(name, domain, labelColumn, set, note) {
+  if (labelColumn !== null) {
+    note?.(
+      CSV_LOSS.RESERVED_NAME,
+      `${domain} column "${name}" is not written: the name is reserved for the label column`,
+      name,
+      set
+    );
+    return false;
+  }
+  note?.(CSV_LOSS.ROLE_ASSUMED, `${domain} column "${name}" reads back with the label role (string)`, name, set);
+  return true;
+}
 function checkRoleColumn(column, role, names, domain, rows, note) {
   if (column === null) {
     return;
@@ -2303,6 +2316,14 @@ function neighbourCell(id, weight) {
   }
   return id.includes(":") ? `${id}:` : id;
 }
+function sourceRunEnd(src, edgeRows, r) {
+  const source = src[edgeRows[r]];
+  let end = r;
+  while (end < edgeRows.length && src[edgeRows[end]] === source) {
+    end++;
+  }
+  return end;
+}
 function* adjacencyLines(snapshot, plan) {
   const { delimiter, newline } = plan.csv;
   const forced = (text) => `"${text.replaceAll('"', '""')}"`;
@@ -2329,10 +2350,7 @@ function* adjacencyLines(snapshot, plan) {
   let r = 0;
   while (r < edgeRows.length) {
     const source = list.src[edgeRows[r]];
-    let end = r;
-    while (end < edgeRows.length && list.src[edgeRows[end]] === source) {
-      end++;
-    }
+    const end = sourceRunEnd(list.src, edgeRows, r);
     const fresh = [];
     let highest = -1;
     let inOrder = true;