blacklight/scribe
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Improve title parsing

Browse source 
The subtitle has been removed because it's difficult to find and error
prone to guess at. It is somewhat accessible from the post's
previewContent field in GraphQL but that can be truncated.
This commit is contained in:
Edward Loveall 2021-10-03 18:14:46 -04:00
parent 2808505b4e
commit fba87c1076
No known key found for this signature in database
GPG key ID: 789A4AE983AC8901
5 changed files with 25 additions and 117 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

66
spec/classes/page_converter_spec.cr
View file

@ -3,56 +3,26 @@ require "../spec_helper"
include Nodes include Nodes
describe PageConverter do describe PageConverter do
it "sets the title and subtitle if present" do it "sets the page title" do
title = "Hello, world!"
paragraph_json = <<-JSON paragraph_json = <<-JSON
[ [
{ {
"text": "Title", "text": "#{title}",
"type": "H3", "type": "H3",
"markups": [], "markups": [],
"iframe": null, "iframe": null,
"layout": null, "layout": null,
"metadata": null "metadata": null
},
{
"text": "Subtitle",
"type": "H4",
"markups": [],
"iframe": null,
"layout": null,
"metadata": null
} }
] ]
JSON JSON
data_json = default_data_json(paragraph_json) data_json = default_data_json(title, paragraph_json)
data = PostResponse::Data.from_json(data_json) data = PostResponse::Data.from_json(data_json)
page = PageConverter.new.convert(data) page = PageConverter.new.convert(data)
page.title.should eq "Title" page.title.should eq title
page.subtitle.should eq "Subtitle"
end
it "sets the title to the first paragraph if no title" do
paragraph_json = <<-JSON
[
{
"text": "Not a title",
"type": "P",
"markups": [],
"iframe": null,
"layout": null,
"metadata": null
}
]
JSON
data_json = default_data_json(paragraph_json)
data = PostResponse::Data.from_json(data_json)
page = PageConverter.new.convert(data)
page.title.should eq "Not a title"
page.subtitle.should eq nil
end end
it "sets the author" do it "sets the author" do
@ -106,25 +76,18 @@ describe PageConverter do
page.created_at.should eq Time.utc(1970, 1, 1, 0, 0, 1) page.created_at.should eq Time.utc(1970, 1, 1, 0, 0, 1)
end end
it "calls ParagraphConverter to convert the remaining paragraph content" do it "calls converts the remaining paragraph content" do
title = "Title"
paragraph_json = <<-JSON paragraph_json = <<-JSON
[ [
{ {
"text": "Title", "text": "#{title}",
"type": "H3", "type": "H3",
"markups": [], "markups": [],
"iframe": null, "iframe": null,
"layout": null, "layout": null,
"metadata": null "metadata": null
}, },
{
"text": "Subtitle",
"type": "H4",
"markups": [],
"iframe": null,
"layout": null,
"metadata": null
},
{ {
"text": "Content", "text": "Content",
"type": "P", "type": "P",
@ -135,7 +98,7 @@ describe PageConverter do
} }
] ]
JSON JSON
data_json = default_data_json(paragraph_json) data_json = default_data_json(title, paragraph_json)
data = PostResponse::Data.from_json(data_json) data = PostResponse::Data.from_json(data_json)
page = PageConverter.new.convert(data) page = PageConverter.new.convert(data)
@ -148,11 +111,18 @@ describe PageConverter do
end end
end end
def default_data_json(paragraph_json : String) def default_paragraph_json
"[]"
end
def default_data_json(
title : String = "This is a story",
paragraph_json : String = default_paragraph_json
)
<<-JSON <<-JSON
{ {
"post": { "post": {
"title": "This is a story", "title": "#{title}",
"createdAt": 1628974309758, "createdAt": 1628974309758,
"creator": { "creator": {
"id": "abc123", "id": "abc123",

20
spec/components/page_content_spec.cr
View file

@ -6,7 +6,6 @@ describe PageContent do
it "renders a single parent/child node structure" do it "renders a single parent/child node structure" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -24,7 +23,6 @@ describe PageContent do
it "renders multiple childrens" do it "renders multiple childrens" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -53,7 +51,6 @@ describe PageContent do
it "renders an anchor" do it "renders an anchor" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -69,7 +66,6 @@ describe PageContent do
it "renders a blockquote" do it "renders a blockquote" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -87,7 +83,6 @@ describe PageContent do
it "renders code" do it "renders code" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -105,7 +100,6 @@ describe PageContent do
it "renders empasis" do it "renders empasis" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -127,7 +121,6 @@ describe PageContent do
children = [Text.new("A caption")] of Child children = [Text.new("A caption")] of Child
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -155,7 +148,6 @@ describe PageContent do
it "renders a GitHub Gist" do it "renders a GitHub Gist" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -173,7 +165,6 @@ describe PageContent do
it "renders an H3" do it "renders an H3" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -191,7 +182,6 @@ describe PageContent do
it "renders an H4" do it "renders an H4" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -209,7 +199,6 @@ describe PageContent do
it "renders an image" do it "renders an image" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -231,7 +220,6 @@ describe PageContent do
it "renders embedded content" do it "renders embedded content" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -256,7 +244,6 @@ describe PageContent do
it "renders an embedded link container" do it "renders an embedded link container" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -280,7 +267,6 @@ describe PageContent do
it "renders an mixtape embed container" do it "renders an mixtape embed container" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -311,7 +297,6 @@ describe PageContent do
it "renders an ordered list" do it "renders an ordered list" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -330,7 +315,6 @@ describe PageContent do
it "renders an preformatted text" do it "renders an preformatted text" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -348,7 +332,6 @@ describe PageContent do
it "renders an preformatted text" do it "renders an preformatted text" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -366,7 +349,6 @@ describe PageContent do
it "renders strong text" do it "renders strong text" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -384,7 +366,6 @@ describe PageContent do
it "renders an unordered list" do it "renders an unordered list" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [
@ -403,7 +384,6 @@ describe PageContent do
it "renders a user anchor" do it "renders a user anchor" do
page = Page.new( page = Page.new(
title: "Title", title: "Title",
subtitle: nil,
author: user_anchor_factory, author: user_anchor_factory,
created_at: Time.local, created_at: Time.local,
nodes: [ nodes: [

51
src/classes/page_converter.cr
View file

@ -1,57 +1,20 @@
struct HeaderData
getter title : String
getter subtitle : String?
def initialize(@title : String, @subtitle : String?)
end
def first_content_paragraph_index : Int32
if title.blank?
0
elsif subtitle.nil? || subtitle.blank?
1
else
2
end
end
end
class PageConverter class PageConverter
def convert(data : PostResponse::Data) : Page def convert(data : PostResponse::Data) : Page
paragraphs = data.post.content.bodyModel.paragraphs title, content = title_and_content(data)
author = data.post.creator author = data.post.creator
created_at = Time.unix_ms(data.post.createdAt) created_at = Time.unix_ms(data.post.createdAt)
header = header_data(paragraphs)
if header.first_content_paragraph_index.zero?
content = [] of PostResponse::Paragraph
else
content = paragraphs[header.first_content_paragraph_index..]
end
Page.new( Page.new(
title: header.title, title: title,
subtitle: header.subtitle,
author: author, author: author,
created_at: Time.unix_ms(data.post.createdAt), created_at: Time.unix_ms(data.post.createdAt),
nodes: ParagraphConverter.new.convert(content) nodes: ParagraphConverter.new.convert(content)
) )
end end
def header_data(paragraphs : Array(PostResponse::Paragraph)) : HeaderData def title_and_content(data : PostResponse::Data) : {String, Array(PostResponse::Paragraph)}
if paragraphs.empty? title = data.post.title
return HeaderData.new("", nil) paragraphs = data.post.content.bodyModel.paragraphs
end non_content_paragraphs = paragraphs.reject { |para| para.text == title }
first_two_paragraphs = paragraphs.first(2) {title, non_content_paragraphs}
first_two_types = first_two_paragraphs.map(&.type)
if first_two_types == [PostResponse::ParagraphType::H3, PostResponse::ParagraphType::H4]
HeaderData.new(
title: first_two_paragraphs[0].text,
subtitle: first_two_paragraphs[1].text,
)
else
HeaderData.new(
title: first_two_paragraphs[0].text,
subtitle: nil,
)
end
end end
end end

2
src/models/page.cr
View file

@ -1,13 +1,11 @@
class Page class Page
getter nodes : Nodes::Children getter nodes : Nodes::Children
getter title : String getter title : String
getter subtitle : String?
getter author : PostResponse::Creator getter author : PostResponse::Creator
getter created_at : Time getter created_at : Time
def initialize( def initialize(
@title : String, @title : String,
@subtitle : String?,
@author : PostResponse::Creator, @author : PostResponse::Creator,
@created_at : Time, @created_at : Time,
@nodes : Nodes::Children = [] of Nodes::Child @nodes : Nodes::Children = [] of Nodes::Child

3
src/pages/articles/show_page.cr
View file

@ -7,9 +7,6 @@ class Articles::ShowPage < MainLayout
def content def content
h1 page.title h1 page.title
if subtitle = page.subtitle
para subtitle, class: "subtitle"
end
para class: "meta" do para class: "meta" do
text "#{author_link(page.author)} on #{page.created_at.to_s("%F")}" text "#{author_link(page.author)} on #{page.created_at.to_s("%F")}"
end end