@graphty/graph-io 0.3.27 → 0.3.29

package/dist/csv.js

@@ -1,8 +1,8 @@
-import { b as plural, bV 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, aL 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, bE as countMixedEdges, $ as agree, e as explicitWeights, b$ as otherFormatDialect, bT 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-CHd4tQE2.js";
-import { i as inferTextDtype, T as TextCellWriter, p as parseTextCell, W as WIDENING_UNSUPPORTED_CODE } from "./chunks/text-DzXjAiTl.js";
+import { b as plural, bV 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, aL 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, bE as countMixedEdges, $ as agree, e as explicitWeights, b$ as otherFormatDialect, bT 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-CVi12U7J.js";
+import { i as inferTextDtype, T as TextCellWriter, p as parseTextCell, W as WIDENING_UNSUPPORTED_CODE } from "./chunks/text-CDgjbkLC.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-CdlvL4RO.js";
-import { o as quoteCsvCell } from "./chunks/escape-D1HPTUrP.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-CdyqwXeY.js";
+import { o as quoteCsvCell } from "./chunks/escape-DQC5HWqF.js";
 import { j as joinText, e as encodeChunks } from "./chunks/writer-DxjgHwSe.js";
 const SOURCE_NAMES = Object.freeze([
   "source",
@@ -1606,7 +1606,7 @@ function sniff(head) {
     label = "utf-16be";
   }
   const body = stripLeadingComments(new TextDecoder(label).decode(head.subarray(0, HEAD_BYTES)), COMMENT_CHARS);
-  if (body.trim().length === 0 || OTHER_FORMATS.some((r) => r.test(body))) {
+  if (body.trim().length === 0 || hasControlCharacter(body) || OTHER_FORMATS.some((r) => r.test(body))) {
     return 0;
   }
   const newline = sniffNewline(body);
@@ -1628,6 +1628,16 @@ function sniff(head) {
   }
   return looksLikeProse(body, newline, delimiter) ? 0 : 0.3;
 }
+function hasControlCharacter(text) {
+  const last = text.length - 1;
+  for (let i = 0; i <= last; i++) {
+    const c = text.codePointAt(i) ?? 0;
+    if (c < 32 && c !== 9 && c !== 10 && c !== 13 && !(c === 26 && i === last)) {
+      return true;
+    }
+  }
+  return false;
+}
 const SENTENCE = /[A-Za-z]{2}[.!?](?:\s|$)/;
 function looksLikeProse(body, newline, delimiter) {
   const lines2 = body.split(newline).filter((l) => l.trim() !== "");
@@ -1901,15 +1911,18 @@ function cellText(column, row) {
     }
   }
 }
-function nodeTableNote(names, table, note) {
+function nodeTableNote(names, table, notes) {
   const written = names.length;
   if (written > 0) {
     const shown = names.slice(0, 3).map((n) => JSON.stringify(n)).join(", ") + (written > 3 ? ", ..." : "");
-    note(
-      CSV_LOSS.NODE_TABLE,
-      `the ${table === "adjacency" ? "adjacency" : "edge"} table has no room for node attributes: ${written} node column${plural(written)} (${shown}) ${agree(written, "is", "are")} written only by a second export with table: "nodes"`,
-      null,
-      written
+    notes.push(
+      Object.freeze({
+        code: CSV_LOSS.NODE_TABLE,
+        message: `the ${table === "adjacency" ? "adjacency" : "edge"} table has no room for node attributes: ${written} node column${plural(written)} (${shown}) ${agree(written, "is", "are")} written only by a second export with table: "nodes"`,
+        column: null,
+        count: written,
+        columns: Object.freeze([...names])
+      })
     );
   }
 }
@@ -1970,7 +1983,7 @@ function planExport(snapshot, options, values) {
       edgeRows.push(e);
     }
   }
-  if (csv.dialect.type === null || csv.table === "adjacency") {
+  if (csv.table !== "nodes" && (csv.dialect.type === null || csv.table === "adjacency")) {
     const where = csv.table === "adjacency" ? "an adjacency table" : "the generic dialect";
     const mixed = countMixedEdges(snapshot);
     if (!snapshot.directed) {
@@ -1996,7 +2009,7 @@ function planExport(snapshot, options, values) {
       0
     );
   }
-  if (folding.mutualCount > 0) {
+  if (folding.mutualCount > 0 && csv.table !== "nodes") {
     note(
       CSV_LOSS.MUTUAL_EXPANDED,
       `${folding.mutualCount} mutual pair${plural(folding.mutualCount)} ${agree(folding.mutualCount, "is", "are")} written as two directed rows; the mutual mark is lost`,
@@ -2067,7 +2080,7 @@ function planExport(snapshot, options, values) {
     nodeTableNote(
       [...nodeLabel === null ? [] : [nodeLabel.meta.name], ...nodeColumns.map((c) => c.column.meta.name)],
       csv.table,
-      note
+      notes
     );
   }
   return {
@@ -2453,7 +2466,8 @@ function headerlessNotes(plan) {
       code: CSV_LOSS.HEADERLESS,
       message: `without a header the file is read back by position (${csv.table === "nodes" ? "id" : "source, target, weight"}, then unnamed columns), so ${names.length === 1 ? "the column" : "the columns"} ${shown} ${agree(names.length, "reads", "read")} back unnamed or in another column's place; write the header, or for an edge list pass dialect: "generic" and write only source, target and weight`,
       column: null,
-      count: names.length
+      count: names.length,
+      columns: Object.freeze([...names])
     })
   ];
 }
@@ -2482,7 +2496,10 @@ function headerlessEdgeColumns(plan) {
 function check(snapshot, options) {
   const plan = planExport(snapshot, options, true);
   return Object.freeze([
-    ...checkCapabilities(snapshot, CSV_CAPABILITIES, plan.common, { roles: KEPT_ROLES }),
+    ...checkCapabilities(snapshot, CSV_CAPABILITIES, plan.common, {
+      roles: KEPT_ROLES,
+      ...plan.csv.table === "nodes" ? { tables: ["node"] } : {}
+    }),
     ...plan.notes,
     ...headerlessNotes(plan)
   ]);