visidata 3.0.2__py3-none-any.whl → 3.1.1__py3-none-any.whl

visidata/interface.py

@@ -10,12 +10,12 @@ vd.theme_option('disp_error_val', '', 'displayed contents for computation except
 vd.theme_option('disp_ambig_width', 1, 'width to use for unicode chars marked ambiguous')
 
 vd.theme_option('disp_pending', '', 'string to display in pending cells')
-vd.theme_option('note_pending', '⌛', 'note to display for pending cells')
-vd.theme_option('note_format_exc', '?', 'cell note for an exception during formatting')
-vd.theme_option('note_getter_exc', '!', 'cell note for an exception during computation')
-vd.theme_option('note_type_exc', '!', 'cell note for an exception during type conversion')
+vd.theme_option('disp_note_pending', ':', 'note to display for pending cells')
+vd.theme_option('disp_note_fmtexc', '?', 'cell note for an exception during formatting')
+vd.theme_option('disp_note_getexc', '!', 'cell note for an exception during computation')
+vd.theme_option('disp_note_typeexc', '!', 'cell note for an exception during type conversion')
 
-vd.theme_option('color_note_pending', 'bold magenta', 'color of note in pending cells')
+vd.theme_option('color_note_pending', 'bold green', 'color of note in pending cells')
 vd.theme_option('color_note_type', '226 yellow', 'color of cell note for non-str types in anytype columns')
 vd.theme_option('color_note_row', '220 yellow', 'color of row note on left edge')
 vd.option('scroll_incr', -3, 'amount to scroll with scrollwheel')
@@ -35,24 +35,23 @@ vd.theme_option('disp_selected_note', '•', '') #
 vd.theme_option('disp_sort_asc', '↑↟⇞⇡⇧⇑', 'characters for ascending sort') # ↑▲↟↥↾↿⇞⇡⇧⇈⤉⤒⥔⥘⥜⥠⍏˄ˆ
 vd.theme_option('disp_sort_desc', '↓↡⇟⇣⇩⇓', 'characters for descending sort') # ↓▼↡↧⇂⇃⇟⇣⇩⇊⤈⤓⥕⥙⥝⥡⍖˅ˇ
 vd.theme_option('color_default', 'white on black', 'the default fg and bg colors')
-vd.theme_option('color_default_hdr', 'bold', 'color of the column headers')
-vd.theme_option('color_bottom_hdr', 'underline', 'color of the bottom header row')
+vd.theme_option('color_default_hdr', 'bold white on black', 'color of the column headers')
+vd.theme_option('color_bottom_hdr', 'underline white on black', 'color of the bottom header row')
 vd.theme_option('color_current_row', 'reverse', 'color of the cursor row')
-vd.theme_option('color_current_col', 'bold', 'color of the cursor column')
+vd.theme_option('color_current_col', 'bold on 232', 'color of the cursor column')
 vd.theme_option('color_current_cell', '', 'color of current cell, if different from color_current_row+color_current_col')
 vd.theme_option('color_current_hdr', 'bold reverse', 'color of the header for the cursor column')
-vd.theme_option('color_column_sep', '246 blue', 'color of column separators')
+vd.theme_option('color_column_sep', 'white on black', 'color of column separators')
 vd.theme_option('color_key_col', '81 cyan', 'color of key columns')
 vd.theme_option('color_hidden_col', '8', 'color of hidden columns on metasheets')
 vd.theme_option('color_selected_row', '215 yellow', 'color of selected rows')
-vd.theme_option('color_clickable', 'underline', 'color of internally clickable item')
+vd.theme_option('color_clickable', 'bold', 'color of internally clickable item')
 vd.theme_option('color_code', 'bold white on 237', 'color of code sample')
-vd.theme_option('color_heading', 'bold 200', 'color of header')
+vd.theme_option('color_heading', 'bold black on yellow', 'color of header')
 vd.theme_option('color_guide_unwritten', '243 on black', 'color of unwritten guides in GuideGuide')
 
 vd.theme_option('force_256_colors', False, 'use 256 colors even if curses reports fewer')
 
 vd.option('quitguard', False, 'confirm before quitting modified sheet')
-vd.option('default_width', 20, 'default column width', replay=True, max_help=1)   # TODO: make not replay and remove from markdown saver
-vd.option('default_height', 4, 'default column height', max_help=-1)
-vd.option('textwrap_cells', True, 'wordwrap text for multiline rows', max_help=1)
+vd.option('default_width', 20, 'default column width', replay=True)   # TODO: make not replay and remove from markdown saver
+vd.option('default_height', 4, 'default column height')

visidata/keys.py

@@ -47,6 +47,9 @@ visidata.vd.prettykeys_trdict = {
         'KEY_SPREVIOUS': 'Shift+PgUp',
         'KEY_SNEXT': 'Shift+PgDn',
 
+        'kxIN': 'FocusIn',
+        'kxOUT': 'FocusOut',
+
         'KEY_BACKSPACE': 'Bksp',
         'BUTTON1_RELEASED': 'LeftBtnUp',
         'BUTTON2_RELEASED': 'MiddleBtnUp',
@@ -72,7 +75,7 @@ for i in range(1, 13):
 
 @visidata.VisiData.api
 def prettykeys(vd, key):
-    if not key:
+    if not key or '+' in key[:-1]:
         return key
 
     for k, v in vd.prettykeys_trdict.items():

visidata/loaders/api_airtable.py

@@ -1,7 +1,7 @@
 import re
 import os
 
-from visidata import vd, date, asyncthread, VisiData, Progress, Sheet, Column, ItemColumn, deduceType, TypedWrapper, setitem
+from visidata import vd, date, asyncthread, VisiData, Progress, Sheet, Column, ItemColumn, deduceType, TypedWrapper, setitem, AttrDict
 
 
 vd.option('airtable_auth_token', '', 'Airtable API key from https://airtable.com/account')

visidata/loaders/archive.py

@@ -15,7 +15,7 @@ def guess_zip(vd, p):
 
 @VisiData.api
 def guess_tar(vd, p):
-    if tarfile.is_tarfile(p.fp):
+    if tarfile.is_tarfile(p.open_bytes()):
         return dict(filetype='tar')
 
 @VisiData.api

visidata/loaders/csv.py

@@ -3,7 +3,9 @@ from visidata import TypedExceptionWrapper, Progress
 
 vd.option('csv_dialect', 'excel', 'dialect passed to csv.reader', replay=True)
 vd.option('csv_delimiter', ',', 'delimiter passed to csv.reader', replay=True)
+vd.option('csv_doublequote', True, 'quote-doubling setting passed to csv.reader', replay=True)
 vd.option('csv_quotechar', '"', 'quotechar passed to csv.reader', replay=True)
+vd.option('csv_quoting', 0, 'quoting style passed to csv.reader and csv.writer', replay=True)
 vd.option('csv_skipinitialspace', True, 'skipinitialspace passed to csv.reader', replay=True)
 vd.option('csv_escapechar', None, 'escapechar passed to csv.reader', replay=True)
 vd.option('csv_lineterminator', '\r\n', 'lineterminator passed to csv.writer', replay=True)
@@ -14,7 +16,10 @@ vd.option('safety_first', False, 'sanitize input/output to handle edge cases, wi
 def guess_csv(vd, p):
     import csv
     csv.field_size_limit(2**31-1)  #288 Windows has max 32-bit
-    line = next(p.open())
+    try:
+        line = next(p.open())
+    except StopIteration:
+        return
     if ',' in line:
         dialect = csv.Sniffer().sniff(line)
         r = dict(filetype='csv', _likelihood=0)
@@ -41,7 +46,7 @@ class CsvSheet(SequenceSheet):
         import csv
         csv.field_size_limit(2**31-1)  #288 Windows has max 32-bit
 
-        with self.open_text_source() as fp:
+        with self.open_text_source(newline='') as fp:
             if options.safety_first:
                 rdr = csv.reader(removeNulls(fp), **options.getall('csv_'))
             else:
@@ -69,11 +74,10 @@ def save_csv(vd, p, sheet):
         if ''.join(colnames):
             cw.writerow(colnames)
 
-        with Progress(gerund='saving'):
+        with Progress(gerund='saving', total=sheet.nRows) as prog:
             for dispvals in sheet.iterdispvals(format=True):
                 cw.writerow(dispvals.values())
-
-CsvSheet.options.regex_skip = '^#.*'
+                prog.addProgress(1)
 
 vd.addGlobals({
     'CsvSheet': CsvSheet

visidata/loaders/eml.py

@@ -14,15 +14,19 @@ class EmailSheet(TableSheet):
         Column('payload', type=vlen, getter=lambda c,r: r.get_payload(decode=False)),
     ]
     def iterload(self):
-        import email
+        import email.parser
         parser = email.parser.Parser()
         with self.source.open(encoding='utf-8') as fp:
             yield from parser.parse(fp).walk()
 
 @EmailSheet.api
 def extract_part(sheet, givenpath, part):
-    with givenpath.open_bytes(mode='w') as fp:
-        fp.write(part.get_payload(decode=True))
+    payload = part.get_payload(decode=True)
+    if payload is None:
+        vd.warning('empty payload')
+    else:
+        with givenpath.open_bytes(mode='w') as fp:
+            fp.write(payload)
 
 @EmailSheet.api
 def extract_parts(sheet, givenpath, *parts):
@@ -37,10 +41,11 @@ def extract_parts(sheet, givenpath, *parts):
         try:
             os.makedirs(givenpath, exist_ok=True)
         except FileExistsError:
-            pass
+            vd.debug(f'{givenpath} already exists')
 
-        for part in parts:
-            vd.execAsync(sheet.extract_part, givenpath / part.get_filename(), part)
+        for i, part in enumerate(parts):
+            fn = part.get_filename() or f'part{i}'
+            vd.execAsync(sheet.extract_part, givenpath / fn, part)
     elif len(parts) == 1:
         vd.execAsync(sheet.extract_part, givenpath, parts[0])
     else:

visidata/loaders/f5log.py

@@ -1092,6 +1092,7 @@ class F5LogSheet(Sheet):
             object_regex = None
 
         try:
+            import zoneinfo
             self._log_tz = zoneinfo.ZoneInfo(
                 vd.options.get("f5log_log_timzeone", "UTC")
             )

visidata/loaders/fec.py

@@ -37,16 +37,14 @@ Thanks to all who have contributed to those projects.
 from copy import copy
 from visidata import (
     vd,
+    VisiData,
     Path,
     Sheet,
     TextSheet,
-    Column,
     ColumnAttr,
     ColumnItem,
-    ENTER,
     asyncthread,
     Progress,
-    addGlobals,
 )
 
 class DiveSheet(Sheet):
@@ -111,10 +109,10 @@ class DiveSheet(Sheet):
                     vd.warning("Can't dive on lists with heterogenous item types.")
                     return False
 
-    def dive(self):
+    def openRow(self, row):
         if self.is_keyvalue:
-            cell = self.cursorRow["value"]
-            name = vd.joinSheetnames(self.name, self.cursorRow["key"])
+            cell = row["value"]
+            name = vd.joinSheetnames(self.name, row["key"])
 
             if isinstance(cell, (list, dict)):
                 vs = self.__class__(name, source = cell)
@@ -123,19 +121,13 @@ class DiveSheet(Sheet):
                 return
         else:
             name = vd.joinSheetnames(self.name, "row")
-            vs = self.__class__(name, source = self.cursorRow)
+            vs = self.__class__(name, source = self.row)
 
         success = vs.reload()
         if success == False:
-            return
-
-        vd.push(vs)
+            vd.fail('could not reload new sheet')
+        return vs
 
-DiveSheet.addCommand(
-    ENTER,
-    'dive-row',
-    'vd.sheet.dive()'
-)
 
 class FECItemizationSheet(Sheet):
     "A sheet to display a list of FEC itemizations from a given form/schedule."
@@ -159,19 +151,9 @@ class FECItemizationSheet(Sheet):
         self.columns.clear()
         for i, name in enumerate(row.keys()):
             self.addColumn(ColumnItem(name))
-    def dive(self):
-        vs = DiveSheet(
-            vd.joinSheetnames(self.name, "detail"),
-            source = self.cursorRow
-        )
-        vs.reload()
-        vd.push(vs)
-
-FECItemizationSheet.addCommand(
-    ENTER,
-    'dive-row',
-    'vd.sheet.dive()'
-)
+
+    def openRow(self, row):
+        return row
 
 class FECScheduleSheet(Sheet):
     "A sheet to display the list of itemized schedules in a filing."
@@ -199,11 +181,8 @@ class FECScheduleSheet(Sheet):
             )
             self.addRow(vs)
 
-FECScheduleSheet.addCommand(
-    ENTER,
-    'dive-row',
-    'vd.push(cursorRow)'
-)
+    def openRow(self, row):
+        return row
 
 COMPONENT_SHEET_CLASSES = {
     "header": DiveSheet,
@@ -230,7 +209,7 @@ class FECFiling(Sheet):
     @asyncthread
     def reload(self):
         from fecfile import fecparser
-        self.rows = []
+        self.rows = []  # rowdef:  Sheet, of a type from COMPONENT_SHEET_CLASSES.values()
 
         row_dict = { }
         itemization_subsheets = {}
@@ -310,16 +289,13 @@ class FECFiling(Sheet):
                 sheet_row.source[form_type].append(item.data)
                 sheet_row.size += 1
 
-FECFiling.addCommand(
-    ENTER,
-    'dive-row',
-    'vd.push(cursorRow)'
-)
+    def openRow(self, row):
+        return row
 
-def open_fec(p):
+@VisiData.api
+def open_fec(vd, p):
     return FECFiling(p.base_stem, source=p)
 
-addGlobals({
-    "open_fec": open_fec,
+vd.addGlobals({
     "DiveSheet": DiveSheet
 })

visidata/loaders/fixed_width.py

@@ -1,5 +1,5 @@
 
-from visidata import VisiData, vd, Sheet, Column, Progress, SequenceSheet
+from visidata import VisiData, vd, Sheet, Column, Progress, SequenceSheet, dispwidth
 
 
 vd.option('fixed_rows', 1000, 'number of rows to check for fixed width columns')
@@ -9,6 +9,22 @@ vd.option('fixed_maxcols', 0, 'max number of fixed-width columns to create (0 is
 def open_fixed(vd, p):
     return FixedWidthColumnsSheet(p.base_stem, source=p, headerlines=[])
 
+@Column.api
+def getMaxDataWidth(col, rows):  #2255 need real max width for fixed width saver
+    '''Return the maximum length of any cell in column or its header,
+    even if wider than window. (Slow for large cells!)'''
+
+    w = 0
+    nlen = dispwidth(col.name)
+    if len(rows) > 0:
+        w_max = 0
+        for r in rows:
+            row_w = dispwidth(col.getDisplayValue(r))
+            if w_max < row_w:
+                w_max = row_w
+        w = w_max
+    return max(w, nlen)
+
 class FixedWidthColumn(Column):
     def __init__(self, name, i, j, **kwargs):
         super().__init__(name, **kwargs)
@@ -38,7 +54,7 @@ def columnize(rows):
     # collapse fields
     for i in allNonspaces:
         if i > prev+1:
-            yield colstart, i
+            yield colstart, prev+1 #2255
             colstart = i
         prev = i
 
@@ -84,7 +100,7 @@ def save_fixed(vd, p, *vsheets):
             widths = {}  # Column -> width:int
             # headers
             for col in Progress(sheet.visibleCols, gerund='sizing'):
-                widths[col] = col.getMaxWidth(sheet.rows)  #1849
+                widths[col] = col.getMaxDataWidth(sheet.rows)  #1849 #2255
                 fp.write(('{0:%s} ' % widths[col]).format(col.name))
             fp.write('\n')
 

visidata/loaders/grep.py

@@ -0,0 +1,121 @@
+#!/usr/bin/python3
+
+from visidata import vd, VisiData, JsonSheet, ColumnAttr, Path, ENTER, AttrDict, ExpectedException, stacktrace, TypedExceptionWrapper
+import json
+import os
+from os import linesep
+
+@VisiData.api
+def open_grep(vd, p):
+    return GrepSheet(p.base_stem, source=p)
+
+@VisiData.api
+def save_grep(vd, p, *vsheets):
+    vd.save_jsonl(p, *vsheets)
+
+def format_row(rowdict):
+    # handle rows that are output of 'rg --json'
+    if 'type' in rowdict and rowdict['type'] == 'match':
+        match_data = rowdict['data']
+        d = {
+            'file':    match_data['path']['text'],
+            'line_no': match_data['line_number'],
+            'text':    match_data['lines']['text'].rstrip(linesep)
+        }
+        return AttrDict(d)
+    # handle a .grep file that was saved by visidata, or
+    # ripgrep rows that were preprocessed by jq:  'rg --json |jq [...]'
+    if 'line_no' in rowdict:
+        rowdict['text'] = rowdict['text'].rstrip(linesep)
+        return AttrDict(rowdict)
+    return None
+
+class GrepSheet(JsonSheet):
+    # The input file is in JSON Lines format, where each line describes a JSON object.
+    # The JSON objects are either in the ripgrep grep_printer format:
+    # https://docs.rs/grep-printer/0.1.0/grep_printer/struct.JSON.html
+    # or contain the keys 'file', 'line_no', and 'text'.
+    _rowtype = 'lines'  # rowdef: AttrDict
+
+    columns = [
+        ColumnAttr('file', type=str),
+        ColumnAttr('line_no', type=int),
+        ColumnAttr('text', type=str)
+    ]
+    nKeys = 2
+    def iterload(self):
+        with self.open_text_source() as fp:
+            for L in fp:
+                try:
+                    if not L: # skip blank lines
+                        continue
+                    json_obj = json.loads(L)
+                    if not isinstance(json_obj, dict):
+                        vd.fail(f'line does not hold a JSON object:  {L}')
+                    row = format_row(json_obj)
+                    if not row: #skip lines that do not contain match data
+                        continue
+                    yield row
+                except ValueError as e:
+                    if self.rows:   # if any rows have been added already
+                        e.stacktrace = stacktrace()
+                        yield TypedExceptionWrapper(json.loads, L, exception=e)  # an error on one line
+                    else:
+                        # If input is not JSON, parse it as output of 'grep -n':  file:line_no:text
+                        # If that does not parse, parse it as output of typical 'grep':  file:text
+                        with self.open_text_source() as fp:
+                            try:
+                                extract_line_no = True
+                                for L in fp:
+                                    L = L.rstrip(linesep)
+                                    sep1 = L.index(':')
+                                    if extract_line_no:
+                                        sep2 = L.find(':', sep1+1)
+                                        try:
+                                            if sep2 == -1: raise ValueError
+                                            line_no = int(L[sep1+1:sep2]) # may raise ValueError
+                                            if line_no < 1: raise ValueError
+                                            text = L[sep2+1:]
+                                        except ValueError: # if we can't find a line_no that is > 0, with a separator after it
+                                            extract_line_no = False
+                                            line_no = None
+                                            text = L[sep1+1:]
+                                    else:
+                                        text = L[sep1+1:]
+                                    yield AttrDict({'file':    L[:sep1],
+                                                    'line_no': line_no,
+                                                    'text':    text})
+                            except ValueError:
+                                vd.fail('file is not grep output')
+                        break
+
+    def afterLoad(self):
+        if self.nRows == 0:
+            vd.status('no grep results found in input data')
+
+@GrepSheet.api
+def sysopen_row(sheet, row):
+    '''Open the file in an editor at the specific line.'''
+    if sheet.nRows == 0: return
+    try:
+        given = row.file
+        if vd.options.grep_base_dir and not os.path.isabs(given):
+            given = vd.options.grep_base_dir + os.sep + row.file
+        p = Path(given)
+    except TypeError:
+        vd.fail(f'cannot open row: {given}')
+    if p.exists():
+        # works for vim and emacsclient
+        if row.line_no is not None:
+            vd.launchEditor(p.given, f'+{row.line_no:d}')
+        else:
+            vd.launchEditor(p.given)
+    else:
+        vd.fail(f'cannot find file: {p.given}')
+
+GrepSheet.addCommand(ENTER, 'sysopen-row', 'sysopen_row(cursorRow)', 'open current file in external $EDITOR, at the line')
+
+vd.addGlobals({
+    'GrepSheet': GrepSheet,
+})
+vd.option('grep_base_dir', None, 'base directory for relative paths opened with sysopen-row')

visidata/loaders/html.py

@@ -2,6 +2,7 @@ import html
 import urllib.parse
 import copy
 import itertools
+import re
 
 from visidata import VisiData, vd, Sheet, options, Column, Progress, IndexSheet, ItemColumn
 

visidata/loaders/http.py

@@ -49,7 +49,12 @@ def openurl_http(vd, path, filetype=None):
         ctx.verify_mode = ssl.CERT_NONE
 
     req = urllib.request.Request(path.given, **vd.options.getall('http_req_'))
-    response = urllib.request.urlopen(req, context=ctx)
+    try:
+        response = urllib.request.urlopen(req, context=ctx)
+    except urllib.error.HTTPError as e:
+        vd.fail(f'cannot open URL: HTTP Error {e.code}: {e.reason}')
+    except urllib.error.URLError as e:
+        vd.fail(f'cannot open URL: {e.reason}')
 
     filetype = filetype or vd.guessFiletype(path, response, funcprefix='guessurl_').get('filetype')  # try guessing by url
     filetype = filetype or vd.guessFiletype(path, funcprefix='guess_').get('filetype')  # try guessing by contents

visidata/loaders/json.py

@@ -1,4 +1,5 @@
 import json
+from collections import Counter
 
 from visidata import vd, date, anytype, VisiData, PyobjSheet, AttrDict, stacktrace, TypedExceptionWrapper, AlwaysDict, ItemColumn, wrapply, TypedWrapper, Progress, Sheet
 
@@ -10,7 +11,10 @@ vd.option('default_colname', '', 'column name to use for non-dict rows')
 @VisiData.api
 def guess_json(vd, p):
     with p.open(encoding=vd.options.encoding) as fp:
-        line = next(fp)
+        try:
+            line = next(fp)
+        except StopIteration:
+            return
 
     line = line.strip()
 
@@ -57,7 +61,7 @@ class JsonSheet(Sheet):
                         yield TypedExceptionWrapper(json.loads, L, exception=e)  # an error on one line
                     else:
                         with self.open_text_source() as fp:
-                            ret = json.load(fp)
+                            ret = json.load(fp, object_hook=AttrDict)
                             if isinstance(ret, list):
                                 yield from ret
                             else:
@@ -81,7 +85,7 @@ class JsonSheet(Sheet):
 
         for k in row:
             if k not in self._knownKeys:
-                c = ItemColumn(k, type=float if isinstance(row[k], (float, int)) else anytype)
+                c = ItemColumn(k)
                 self.addColumn(c)
 
         return ret
@@ -90,7 +94,7 @@ class JsonSheet(Sheet):
         return AttrDict(fields)
 
     def openRow(self, row):
-        return PyobjSheet("%s[%s]" % (self.name, self.keystr(row)), source=row)
+        return PyobjSheet("%s[%s]" % (self.name, self.rowname(row)), source=row)
 
 JsonSheet.init('_knownKeys', set, copy=True)  # set of row keys already seen
 
@@ -139,6 +143,13 @@ def save_json(vd, p, *vsheets):
 
         jsonenc = _vjsonEncoder(indent=indent, sort_keys=vs.options.json_sort_keys, ensure_ascii=vs.options.json_ensure_ascii)
 
+        dupnames = find_duplicates([vs.name for vs in vsheets])
+        for name in dupnames:
+            vd.warning('json cannot save sheet with duplicated name: ' + name)
+        for vs in vsheets:
+            dupnames = find_duplicates([c.name for c in vs.visibleCols])
+            for name in dupnames:
+                vd.warning('json cannot save column with duplicated name: ' + name)
         if len(vsheets) == 1:
             fp.write('[\n')
             vs = vsheets[0]
@@ -161,6 +172,9 @@ def save_json(vd, p, *vsheets):
 def write_jsonl(vs, fp):
         vcols = vs.visibleCols
         jsonenc = _vjsonEncoder()
+        dupnames = find_duplicates([c.name for c in vcols])
+        for name in dupnames:
+            vd.warning('json cannot save column with duplicated name: ' + name)
         with Progress(gerund='saving'):
             for i, row in enumerate(vs.iterrows()):
                 rowdict = _rowdict(vcols, row, keep_nulls=(i==0))
@@ -177,6 +191,8 @@ def write_jsonl(vs, fp):
 @VisiData.api
 def save_jsonl(vd, p, *vsheets):
     with p.open(mode='w', encoding=vsheets[0].options.save_encoding) as fp:
+        if len(vsheets) > 1:
+            vd.warning('jsonl cannot separate sheets yet. Concatenating all rows.')
         for vs in vsheets:
             vs.write_jsonl(fp)
 
@@ -186,6 +202,8 @@ def JSON(vd, s:str):
     'Parse `s` as JSON.'
     return json.loads(s)
 
+def find_duplicates(names):
+    return list(colname for colname,count in Counter(names).items() if count > 1)
 
 JsonSheet.options.encoding = 'utf-8'
 JsonSheet.options.regex_skip = r'^(//|#).*'

visidata/loaders/jsonla.py

@@ -18,10 +18,16 @@ def guess_jsonla(vd, p):
     '''
 
     with p.open(encoding=vd.options.encoding) as fp:
-        first_line = next(fp)
+        try:
+            first_line = next(fp)
+        except StopIteration:
+            return
 
     if first_line.strip().startswith('['):
-        ret = json.loads(first_line)
+        try:
+            ret = json.loads(first_line)
+        except json.decoder.JSONDecodeError:
+            return
         if isinstance(ret, list) and all(isinstance(v, str) for v in ret):
             return dict(filetype='jsonla')
 

visidata/loaders/mailbox.py

@@ -4,6 +4,7 @@ from visidata import VisiData, Sheet, ItemColumn, date, Column
 @VisiData.api
 def open_mbox(vd, p):
     return MboxSheet(p.base_stem, source=p, format='mbox')
+VisiData.open_mailbox = VisiData.open_mbox
 
 @VisiData.api
 def open_maildir(vd, p):

visidata/loaders/markdown.py

@@ -1,5 +1,11 @@
 from visidata import VisiData, vd, options, Progress
 
+def markdown_link(s, href):
+    if not href:
+        return s
+
+    return f'[{s}]({href})'
+
 def markdown_escape(s, style='orgmode'):
     if style == 'jira':
         return s
@@ -31,17 +37,30 @@ def write_md(p, *vsheets, md_style='orgmode'):
             if len(vsheets) > 1:
                 fp.write('# %s\n\n' % vs.name)
 
-            fp.write(delim + delim.join('%-*s' % (col.width or options.default_width, markdown_escape(col.name, md_style)) for col in vs.visibleCols) + '|\n')
+            hdrs = []
+            for col in vs.visibleCols:
+                if col.name.endswith('_href'):
+                    continue
+                hdrs.append('%-*s' % (col.width or options.default_width, markdown_escape(col.name, md_style)))
+
+            fp.write(delim + delim.join(hdrs) + delim + '\n')
+
             if md_style == 'orgmode':
-                fp.write('|' + '|'.join(markdown_colhdr(col) for col in vs.visibleCols) + '|\n')
+                fp.write('|' + '|'.join(markdown_colhdr(col) for col in vs.visibleCols if not col.name.endswith('_href')) + '|\n')
 
             with Progress(gerund='saving'):
                 for dispvals in vs.iterdispvals(format=True):
-                    s = '|'
+                    vals = []
                     for col, val in dispvals.items():
-                        s += '%-*s|' % (col.width or options.default_width, markdown_escape(val, md_style))
-                    s += '\n'
-                    fp.write(s)
+                        if col.name.endswith('_href'):
+                            continue
+                        val = markdown_escape(val, md_style)
+                        linkcol = vs.colsByName.get(col.name + '_href')
+                        if linkcol:
+                            val = markdown_link(val, dispvals.get(linkcol))
+                        vals.append('%-*s' % (col.width or options.default_width, val))
+                    fp.write('|' + '|'.join(vals) + '|\n')
+
             fp.write('\n')