@zentered/issue-forms-body-parser 1.2.1 → 1.4.1

package/src/parse.js

@@ -5,139 +5,88 @@ import remarkParse from 'remark-parse'
 import remarkGfm from 'remark-gfm'
 import slugify from '@sindresorhus/slugify'
 import remarkStringify from 'remark-stringify'
-import { parse, isMatch } from 'date-fns'
 import stripFinalNewline from 'strip-final-newline'
 
-// if the system time is not UTC, we need to convert it to UTC
-import { zonedTimeToUtc, formatInTimeZone } from 'date-fns-tz/esm'
-const loc = 'UTC'
+import {
+  parseDate,
+  parseTime,
+  parseDuration,
+  parseList
+} from './parsers/index.js'
 
-const commonDateFormats = [
-  'yyyy-MM-dd',
-  'dd/MM/yyyy',
-  'dd/MM/yy',
-  'dd-MM-yyyy',
-  'dd-MM-yy',
-  'dd.MM.yyyy',
-  'dd.MM.yy'
-]
+export default async function parseMD(body) {
+  const tokens = await unified().use(remarkParse).use(remarkGfm).parse(body)
+  if (!tokens) {
+    return []
+  }
 
-const commonTimeFormats = ['HH:mm', 'HH.mm', 'hh:mm a', 'hh:mm A']
+  const structuredResponse = {}
+  let currentHeading = null
+  for (const token of tokens.children) {
+    const text = await unified()
+      .use(remarkGfm)
+      .use(remarkStringify)
+      .stringify(token)
+    const cleanText = stripFinalNewline(text)
 
-function parseDuration(text) {
-  const duration = {
-    hours: 0,
-    minutes: 0
-  }
+    // issue forms uses h3 as a heading
+    if (token.type === 'heading' && token.depth === 3) {
+      currentHeading = slugify(token.children[0].value)
+      structuredResponse[currentHeading] = {
+        title: token.children[0].value,
+        content: []
+      }
+    } else if (token.type === 'paragraph' && currentHeading) {
+      const obj = structuredResponse[currentHeading]
 
-  const pieces = text.replace('m', '').split('h')
-  duration.hours = parseInt(pieces[0]) ? parseInt(pieces[0]) : 0
-  duration.minutes = parseInt(pieces[1]) ? parseInt(pieces[1]) : 0
-  return duration
-}
+      const date = parseDate(cleanText)
+      const time = parseTime(cleanText)
+      const duration = parseDuration(cleanText)
 
-function parseDate(text) {
-  const match = commonDateFormats.map((format) => {
-    return isMatch(text, format)
-  })
-  if (match.indexOf(true) > -1) {
-    const date = zonedTimeToUtc(
-      parse(text, commonDateFormats[match.indexOf(true)], new Date()),
-      loc
-    ).toJSON()
-    return date.split('T')[0]
-  } else {
-    return null
-  }
-}
+      if (date) {
+        obj.date = date
+      }
 
-function parseTime(text) {
-  const match = commonTimeFormats.map((format) => {
-    return isMatch(text, format)
-  })
-  if (match.indexOf(true) > -1) {
-    const time = zonedTimeToUtc(
-      parse(text, commonTimeFormats[match.indexOf(true)], new Date()),
-      loc
-    )
-    return formatInTimeZone(time, loc, 'HH:mm')
-  } else {
-    return null
-  }
-}
+      if (time) {
+        obj.time = time
+      }
 
-function parseList(list) {
-  return list.children
-    .map((item) => {
-      const listItem = {}
-      if (item.type === 'list') {
-        return parseList(list)
-      } else if (item.type === 'listItem') {
-        listItem.checked = item.checked
-        return item.children
-          .map((child) => {
-            if (child.type === 'paragraph') {
-              listItem.text = child.children
-                .map((c) => {
-                  if (c.type === 'link') {
-                    return c.children[0].value
-                  } else {
-                    return c.value
-                  }
-                })
-                .filter((x) => !!x)
-                .join('')
-              return listItem
-            }
-          })
-          .filter((x) => !!x)
+      if (duration) {
+        obj.duration = duration
       }
-    })
-    .filter((x) => !!x)
-}
 
-export default async function parseMD(body) {
-  const tokens = await unified().use(remarkParse).use(remarkGfm).parse(body)
-  if (!tokens) {
-    return []
+      obj.content.push(cleanText)
+    } else if (token.type === 'list') {
+      const obj = structuredResponse[currentHeading]
+      obj.text = cleanText
+      obj.list = parseList(token).flat()
+    } else if (token.type === 'html') {
+      const obj = structuredResponse[currentHeading]
+      obj.content.push(token.html)
+    } else if (token.type === 'code') {
+      const obj = structuredResponse[currentHeading]
+      obj.lang = token.lang
+      obj.text = cleanText
+    } else if (token.type === 'heading' && token.depth > 3) {
+      const obj = structuredResponse[currentHeading]
+      obj.content.push(token.children[0].value)
+    } else {
+      console.log('unhandled token type')
+      console.log(token)
+    }
   }
 
-  const r = []
-  for (let idx = 0; idx < tokens.children.length; idx = idx + 2) {
-    const current = tokens.children[idx]
-    const hasNext = idx + 1 < tokens.children.length
-
-    if (current.type === 'heading') {
-      // issue-form answers start with a h3 heading, ignore everything else
-      const obj = {
-        id: slugify(current.children[0].value),
-        title: current.children[0].value
-      }
-      if (hasNext) {
-        const next = tokens.children[idx + 1]
-        if (next.type === 'list') {
-          obj.list = parseList(next).flat()
-        }
-        const text = await unified()
-          .use(remarkGfm)
-          .use(remarkStringify)
-          .stringify(next)
-        obj.text = stripFinalNewline(text)
-        const date = parseDate(obj.text)
-        const time = parseTime(obj.text)
-        if (date) {
-          obj.date = date
-        }
-        if (time) {
-          obj.time = time
-        }
-        if (obj.id === 'duration') {
-          obj.duration = parseDuration(obj.text)
-        }
+  for (const key in structuredResponse) {
+    const token = structuredResponse[key]
+    const content = token.content.filter(Boolean)
+    if (content && content.length > 0) {
+      if (content.length === 1) {
+        token.text = content[0]
       }
-      r.push(obj)
+      token.text = content.join('\n\n')
     }
+    token.content = content
   }
 
-  return r
+  return structuredResponse
 }

package/src/parsers/date.js

@@ -0,0 +1,30 @@
+'use strict'
+
+import { parse, isMatch } from 'date-fns'
+import { zonedTimeToUtc } from 'date-fns-tz/esm'
+
+const loc = 'UTC'
+const commonDateFormats = [
+  'yyyy-MM-dd',
+  'dd/MM/yyyy',
+  'dd/MM/yy',
+  'dd-MM-yyyy',
+  'dd-MM-yy',
+  'dd.MM.yyyy',
+  'dd.MM.yy'
+]
+
+export default function parseDate(text) {
+  const match = commonDateFormats.map((format) => {
+    return isMatch(text, format)
+  })
+  if (match.indexOf(true) > -1) {
+    const date = zonedTimeToUtc(
+      parse(text, commonDateFormats[match.indexOf(true)], new Date()),
+      loc
+    ).toJSON()
+    return date.split('T')[0]
+  } else {
+    return null
+  }
+}

package/src/parsers/duration.js

@@ -0,0 +1,30 @@
+'use strict'
+
+export default function parseDuration(text) {
+  let matched = false
+  const duration = {
+    hours: 0,
+    minutes: 0
+  }
+
+  const hoursAndMinutes = new RegExp(/([0-9]+)h([0-9]+)m/)
+  const hours = new RegExp(/([0-9]+)h/)
+
+  if (text.match(hoursAndMinutes)) {
+    matched = true
+    const [, h, m] = text.match(hoursAndMinutes)
+    duration.hours = parseInt(h)
+    duration.minutes = parseInt(m)
+  } else if (text.match(hours)) {
+    matched = true
+    const [, h] = text.match(hours)
+    duration.hours = parseInt(h)
+    duration.minutes = 0
+  }
+
+  if (matched) {
+    return duration
+  } else {
+    return null
+  }
+}

package/src/parsers/index.js

@@ -0,0 +1,11 @@
+'use strict'
+
+import date from './date.js'
+import time from './time.js'
+import duration from './duration.js'
+import list from './list.js'
+
+export const parseDate = date
+export const parseTime = time
+export const parseDuration = duration
+export const parseList = list

package/src/parsers/list.js

@@ -0,0 +1,31 @@
+'use strict'
+
+export default function parseList(list) {
+  return list.children
+    .map((item) => {
+      const listItem = {}
+      if (item.type === 'list') {
+        return parseList(list)
+      } else if (item.type === 'listItem') {
+        listItem.checked = item.checked
+        return item.children
+          .map((child) => {
+            if (child.type === 'paragraph') {
+              listItem.text = child.children
+                .map((c) => {
+                  if (c.type === 'link') {
+                    return c.children[0].value
+                  } else {
+                    return c.value
+                  }
+                })
+                .filter((x) => !!x)
+                .join('')
+              return listItem
+            }
+          })
+          .filter((x) => !!x)
+      }
+    })
+    .filter((x) => !!x)
+}

package/src/parsers/time.js

@@ -0,0 +1,22 @@
+'use strict'
+
+import { parse, isMatch } from 'date-fns'
+import { zonedTimeToUtc, formatInTimeZone } from 'date-fns-tz/esm'
+
+const loc = 'UTC'
+const commonTimeFormats = ['HH:mm', 'HH.mm', 'hh:mm a', 'hh:mm A']
+
+export default function parseTime(text) {
+  const match = commonTimeFormats.map((format) => {
+    return isMatch(text, format)
+  })
+  if (match.indexOf(true) > -1) {
+    const time = zonedTimeToUtc(
+      parse(text, commonTimeFormats[match.indexOf(true)], new Date()),
+      loc
+    )
+    return formatInTimeZone(time, loc, 'HH:mm')
+  } else {
+    return null
+  }
+}

package/test/parse-issue.test.js

@@ -8,84 +8,78 @@ import { join } from 'path'
 const test = t.test
 
 test('parse(md) should parse GitHub Issue Form data into useful, structured data', async (t) => {
-  const expected = [
-    {
-      id: 'event-description',
+  const expected = {
+    'event-description': {
       title: 'Event Description',
-      text: "Let's meet for coffee and chat about tech, coding, Cyprus and the newly formed\nCDC (Cyprus Developer Community)."
+      content: [
+        'Welcome to the CDC - Cyprus Developer Community! Join us for our monthly Larnaka\nmeet & greet event. Meet likeminded people, discuss topics we would like to hear\nabout in upcoming talks, welcome potential speakers, discuss all things tech and\nhave fun!',
+        'Notice with regards to COVID:',
+        'All attendees must follow measures in accordance with Ministry of Health\ndirectives. <https://www.pio.gov.cy/coronavirus/eng>'
+      ],
+      text: 'Welcome to the CDC - Cyprus Developer Community! Join us for our monthly Larnaka\nmeet & greet event. Meet likeminded people, discuss topics we would like to hear\nabout in upcoming talks, welcome potential speakers, discuss all things tech and\nhave fun!\n\nNotice with regards to COVID:\n\nAll attendees must follow measures in accordance with Ministry of Health\ndirectives. <https://www.pio.gov.cy/coronavirus/eng>'
     },
-    {
-      id: 'location',
+    location: {
       title: 'Location',
+      content: [
+        '[Cafe Nero Finikoudes, Larnaka](https://goo.gl/maps/Bzjxdeat3BSdsUSVA)'
+      ],
       text: '[Cafe Nero Finikoudes, Larnaka](https://goo.gl/maps/Bzjxdeat3BSdsUSVA)'
     },
-    {
-      id: 'date',
+    date: {
       title: 'Date',
-      text: '11.03.2022',
-      date: '2022-03-11'
+      content: ['11.03.2022'],
+      date: '2022-03-11',
+      text: '11.03.2022'
     },
-    { id: 'time', title: 'Time', text: '16:00', time: '16:00' },
-    {
-      id: 'duration',
+    time: { title: 'Time', content: ['16:00'], time: '16:00', text: '16:00' },
+    duration: {
       title: 'Duration',
-      text: '2h',
-      duration: { hours: 2, minutes: 0 }
+      content: ['2h'],
+      duration: { hours: 2, minutes: 0 },
+      text: '2h'
     },
-    {
-      id: 'list-item-checked',
+    'list-item-checked': {
       title: 'List Item Checked',
+      content: [],
+      text: "*   [x] I agree to follow this project's\n    [Code of Conduct](https://berlincodeofconduct.org)",
       list: [
         {
           checked: true,
           text: "I agree to follow this project's\nCode of Conduct"
         }
-      ],
-      text: "*   [x] I agree to follow this project's\n    [Code of Conduct](https://berlincodeofconduct.org)"
+      ]
     },
-    {
-      id: 'list-item-unchecked',
+    'list-item-unchecked': {
       title: 'List Item Unchecked',
+      content: [],
+      text: "*   [ ] I agree to follow this project's\n    [Code of Conduct](https://berlincodeofconduct.org)",
       list: [
         {
           checked: false,
           text: "I agree to follow this project's\nCode of Conduct"
         }
-      ],
-      text: "*   [ ] I agree to follow this project's\n    [Code of Conduct](https://berlincodeofconduct.org)"
+      ]
     },
-    {
-      id: 'mixed-task-list',
+    'mixed-task-list': {
       title: 'Mixed Task List',
+      content: [],
+      text: '*   [x] checked\n*   [ ] unchecked\n*   [x] checked 2\n*   [x] checked 3\n*   [ ] unchecked 2',
       list: [
         { checked: true, text: 'checked' },
         { checked: false, text: 'unchecked' },
         { checked: true, text: 'checked 2' },
         { checked: true, text: 'checked 3' },
         { checked: false, text: 'unchecked 2' }
-      ],
-      text: '*   [x] checked\n*   [ ] unchecked\n*   [x] checked 2\n*   [x] checked 3\n*   [ ] unchecked 2'
+      ]
     },
-    {
-      id: 'complex-list',
-      title: 'Complex List',
-      list: [
-        { checked: null, text: 'one' },
-        { checked: null, text: 'two' }
-      ],
-      text: '*   one\n*   two\n    *   three\n    *   four\n        1.  five\n        2.  six'
-    },
-    {
-      id: 'repositories',
+    repositories: {
       title: 'Repositories',
+      content: [],
+      lang: 'csv',
       text: '```csv\nhttps://example.com/repository-1\nhttps://example.com/repository-2\n```'
     },
-    {
-      id: 'visibility',
-      title: 'Visibility',
-      text: 'Internal'
-    }
-  ]
+    visibility: { title: 'Visibility', content: ['Internal'], text: 'Internal' }
+  }
 
   const md = await readFile(
     join(process.cwd(), 'test', 'test-issue-1.md'),
@@ -93,11 +87,12 @@ test('parse(md) should parse GitHub Issue Form data into useful, structured data
   )
   const actual = await fn(md)
   // console.log(JSON.stringify(actual, null, 0))
-  t.deepEqual(actual, expected)
+
+  t.same(actual, expected)
 })
 
 test('parse(md) return nothing', async (t) => {
-  const expected = []
+  const expected = {}
 
   const md = await readFile(
     join(process.cwd(), 'test', 'test-issue-2.md'),
@@ -105,5 +100,5 @@ test('parse(md) return nothing', async (t) => {
   )
   const actual = await fn(md)
   // console.log(JSON.stringify(actual, null, 0))
-  t.deepEqual(actual, expected)
+  t.same(actual, expected)
 })

package/test/test-issue-1.md

@@ -1,7 +1,16 @@
 ### Event Description
 
-Let's meet for coffee and chat about tech, coding, Cyprus and the newly formed
-CDC (Cyprus Developer Community).
+<img width="1107" alt="The Brewery Larnaka" src="https://user-images.githubusercontent.com/3581331/162574191-3c023b32-34d9-4035-90bf-7297cdccaf06.png">
+
+Welcome to the CDC - Cyprus Developer Community! Join us for our monthly Larnaka
+meet & greet event. Meet likeminded people, discuss topics we would like to hear
+about in upcoming talks, welcome potential speakers, discuss all things tech and
+have fun!
+
+#### Notice with regards to COVID:
+
+All attendees must follow measures in accordance with Ministry of Health
+directives. https://www.pio.gov.cy/coronavirus/eng
 
 ### Location
 
@@ -37,15 +46,6 @@ CDC (Cyprus Developer Community).
 - [x] checked 3
 - [ ] unchecked 2
 
-### Complex List
-
-- one
-- two
-  - three
-  - four
-    1. five
-    2. six
-
 ### Repositories
 
 ```csv