firecrawl-sdk 1.12.0 → 1.14.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 361f6ab9151b9481824965c07081793bb5a21aa99ef9b5e8ad61ff3db58b64e8
|
|
4
|
+
data.tar.gz: d99e0a59c24b8cda8618a0d2fa83225c3869dea31768eb9846e4dad7173ab305
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 145c0ec698d497bbf326157cb304dee1c4e1d42439865d247b60fcb5ff4fa4b7ebcf818ac86dd0048ca192690570ff501052d65f4254464044d18c6e71b22c61
|
|
7
|
+
data.tar.gz: 2a2010f8e0b1545d662d782093bfad674be3f1406811ec330e00da0b2addedcd0a9a8e9ccd0821e0c48894e43acf6b72c0beb377e623ba23000b09e7612f8790
|
|
@@ -8,7 +8,7 @@ module Firecrawl
|
|
|
8
8
|
:metadata, :links, :images, :screenshot, :audio,
|
|
9
9
|
:video, :attributes, :actions, :answer, :highlights,
|
|
10
10
|
:warning, :change_tracking, :branding, :product, :menu,
|
|
11
|
-
:blocks
|
|
11
|
+
:pages, :blocks
|
|
12
12
|
|
|
13
13
|
def initialize(data)
|
|
14
14
|
@markdown = data["markdown"]
|
|
@@ -31,6 +31,7 @@ module Firecrawl
|
|
|
31
31
|
@branding = data["branding"]
|
|
32
32
|
@product = data["product"] && ProductProfile.new(data["product"])
|
|
33
33
|
@menu = data["menu"] && MenuProfile.new(data["menu"])
|
|
34
|
+
@pages = data["pages"]
|
|
34
35
|
@blocks = data["blocks"]
|
|
35
36
|
end
|
|
36
37
|
|
|
@@ -35,7 +35,7 @@ module Firecrawl
|
|
|
35
35
|
"excludeTags" => exclude_tags,
|
|
36
36
|
"onlyMainContent" => only_main_content,
|
|
37
37
|
"timeout" => timeout,
|
|
38
|
-
"parsers" => parsers,
|
|
38
|
+
"parsers" => parsers&.map { |parser| parser.respond_to?(:to_h) ? parser.to_h : parser },
|
|
39
39
|
"skipTlsVerification" => skip_tls_verification,
|
|
40
40
|
"removeBase64Images" => remove_base64_images,
|
|
41
41
|
"blockAds" => block_ads,
|
|
@@ -0,0 +1,33 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
module Firecrawl
|
|
4
|
+
module Models
|
|
5
|
+
# PDF parser configuration for use in ScrapeOptions / ParseOptions parsers.
|
|
6
|
+
class PDFParser
|
|
7
|
+
attr_reader :mode, :max_pages, :pages, :blocks, :page_markers
|
|
8
|
+
|
|
9
|
+
def initialize(mode: nil, max_pages: nil, pages: nil, blocks: nil, page_markers: nil)
|
|
10
|
+
@mode = mode
|
|
11
|
+
@max_pages = max_pages
|
|
12
|
+
@pages = pages
|
|
13
|
+
@blocks = blocks
|
|
14
|
+
@page_markers = page_markers
|
|
15
|
+
end
|
|
16
|
+
|
|
17
|
+
def to_h
|
|
18
|
+
{
|
|
19
|
+
"type" => "pdf",
|
|
20
|
+
"mode" => mode,
|
|
21
|
+
"maxPages" => max_pages,
|
|
22
|
+
"pages" => pages,
|
|
23
|
+
"blocks" => blocks,
|
|
24
|
+
"pageMarkers" => page_markers,
|
|
25
|
+
}.compact
|
|
26
|
+
end
|
|
27
|
+
|
|
28
|
+
def type
|
|
29
|
+
"pdf"
|
|
30
|
+
end
|
|
31
|
+
end
|
|
32
|
+
end
|
|
33
|
+
end
|
|
@@ -31,7 +31,7 @@ module Firecrawl
|
|
|
31
31
|
"timeout" => timeout,
|
|
32
32
|
"waitFor" => wait_for,
|
|
33
33
|
"mobile" => mobile,
|
|
34
|
-
"parsers" => parsers,
|
|
34
|
+
"parsers" => parsers&.map { |parser| parser.respond_to?(:to_h) ? parser.to_h : parser },
|
|
35
35
|
"actions" => actions,
|
|
36
36
|
"location" => location.is_a?(Hash) ? location : location&.to_h,
|
|
37
37
|
"skipTlsVerification" => skip_tls_verification,
|
data/lib/firecrawl/version.rb
CHANGED
data/lib/firecrawl.rb
CHANGED
|
@@ -4,6 +4,7 @@ require_relative "firecrawl/version"
|
|
|
4
4
|
require_relative "firecrawl/errors"
|
|
5
5
|
require_relative "firecrawl/http_client"
|
|
6
6
|
require_relative "firecrawl/models/query_format"
|
|
7
|
+
require_relative "firecrawl/models/pdf_parser"
|
|
7
8
|
require_relative "firecrawl/models/product_profile"
|
|
8
9
|
require_relative "firecrawl/models/menu_profile"
|
|
9
10
|
require_relative "firecrawl/models/document"
|
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: firecrawl-sdk
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 1.
|
|
4
|
+
version: 1.14.0
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Firecrawl
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2026-08-
|
|
11
|
+
date: 2026-08-21 00:00:00.000000000 Z
|
|
12
12
|
dependencies: []
|
|
13
13
|
description: A type-safe Ruby client for the Firecrawl v2 API. Supports scraping,
|
|
14
14
|
crawling, batch scraping, URL mapping, web search, and AI agent operations.
|
|
@@ -43,6 +43,7 @@ files:
|
|
|
43
43
|
- lib/firecrawl/models/monitor.rb
|
|
44
44
|
- lib/firecrawl/models/parse_file.rb
|
|
45
45
|
- lib/firecrawl/models/parse_options.rb
|
|
46
|
+
- lib/firecrawl/models/pdf_parser.rb
|
|
46
47
|
- lib/firecrawl/models/product_profile.rb
|
|
47
48
|
- lib/firecrawl/models/query_format.rb
|
|
48
49
|
- lib/firecrawl/models/scrape_options.rb
|