@opentermsarchive/engine 1.1.2 → 1.2.0

package/src/archivist/extract/index.test.js

@@ -4,11 +4,13 @@ import { fileURLToPath } from 'url';
 
 import chai from 'chai';
 import jsdom from 'jsdom';
+import mime from 'mime';
 
-import { InaccessibleContentError } from '../errors.js';
 import SourceDocument from '../services/sourceDocument.js';
 
-import { convertRelativeURLsToAbsolute, extractFromHTML, extractFromPDF } from './index.js';
+import { ExtractDocumentError } from './errors.js';
+
+import extract, { convertRelativeURLsToAbsolute } from './index.js';
 
 const __dirname = path.dirname(fileURLToPath(import.meta.url));
 const fs = fsApi.promises;
@@ -28,6 +30,7 @@ const rawHTML = `
     <p><a id="link1" href="/relative/link">link 1</a></p>
     <p><a id="link2" href="#anchor">link 2</a></p>
     <p><a id="link3" href="http://absolute.url/link">link 3</a></p>
+    <p><a id="link5" href="http://[INVALID_URL=http://www.example.org/">link 5</a></p>
     <div id="empty"></div>
     <div id="whitespaceOnly"> </div>
   </body>
@@ -40,7 +43,9 @@ const expectedExtracted = `Title
 
 [link 2](#anchor)
 
-[link 3](http://absolute.url/link)`;
+[link 3](http://absolute.url/link)
+
+[link 5](http://[INVALID_URL=http://www.example.org/)`;
 
 const expectedExtractedWithAdditional = `Title
 =====`;
@@ -92,7 +97,7 @@ const additionalFilter = {
       link.remove();
     });
   },
-  removeLinksAsync: async function removeLinksAsync(document) {
+  removeLinksAsync: function removeLinksAsync(document) {
     return new Promise(resolve => {
       setTimeout(() => {
         const links = document.querySelectorAll('a');
@@ -117,400 +122,412 @@ describe('Extract', () => {
       subject = Array.from(webPageDOM.querySelectorAll('a[href]')).map(el => el.href);
     });
 
-    it('converts relative urls', async () => {
+    it('converts relative urls', () => {
       expect(subject).to.include('https://exemple.com/relative/link');
     });
 
-    it('leaves absolute urls untouched', async () => {
+    it('leaves absolute urls untouched', () => {
       expect(subject).to.include('http://absolute.url/link');
     });
-  });
 
-  describe('#extractFromHTML', () => {
-    describe('Select', () => {
-      context('with string selector', () => {
-        it('extracts content from the given HTML with common changing items', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            location: virtualLocation,
-            contentSelectors: 'body',
-            content: rawHTMLWithCommonChangingItems,
-          }));
-
-          expect(result).to.equal(expectedExtractedWithCommonChangingItems);
-        });
+    it('leaves invalid urls untouched', () => {
+      expect(subject).to.include('http://[INVALID_URL=http://www.example.org/');
+    });
+  });
 
-        it('extracts content from the given HTML', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            location: virtualLocation,
-            contentSelectors: 'body',
-            content: rawHTML,
-          }));
+  describe('#extract', () => {
+    context('from HTML content', () => {
+      describe('Select', () => {
+        context('with string selector', () => {
+          it('extracts content from the given HTML with common changing items', async () => {
+            const result = await extract(new SourceDocument({
+              location: virtualLocation,
+              contentSelectors: 'body',
+              content: rawHTMLWithCommonChangingItems,
+            }));
 
-          expect(result).to.equal(expectedExtracted);
-        });
+            expect(result).to.equal(expectedExtractedWithCommonChangingItems);
+          });
 
-        context('with no match for the given selector', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
+          it('extracts content from the given HTML', async () => {
+            const result = await extract(new SourceDocument({
               location: virtualLocation,
-              contentSelectors: '#thisAnchorDoesNotExist',
+              contentSelectors: 'body',
               content: rawHTML,
-            }))).to.be.rejectedWith(InaccessibleContentError, /#thisAnchorDoesNotExist/);
+            }));
+
+            expect(result).to.equal(expectedExtracted);
           });
-        });
 
-        context('with no content for the matching given selector', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
-              location: virtualLocation,
-              contentSelectors: '#empty',
-              content: rawHTML,
-            }))).to.be.rejectedWith(InaccessibleContentError, /empty content/);
+          context('with no match for the given selector', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                location: virtualLocation,
+                contentSelectors: '#thisAnchorDoesNotExist',
+                content: rawHTML,
+              }))).to.be.rejectedWith(Error, /#thisAnchorDoesNotExist/);
+            });
           });
-        });
 
-        context('with a whitespace only content for the corresponding given selector', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
-              location: virtualLocation,
-              contentSelectors: '#whitespaceOnly',
-              content: rawHTML,
-            }))).to.be.rejectedWith(InaccessibleContentError, /empty content/);
+          context('with no content for the matching given selector', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                location: virtualLocation,
+                contentSelectors: '#empty',
+                content: rawHTML,
+              }))).to.be.rejectedWith(ExtractDocumentError, /empty content/);
+            });
+          });
+
+          context('with a whitespace only content for the corresponding given selector', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                location: virtualLocation,
+                contentSelectors: '#whitespaceOnly',
+                content: rawHTML,
+              }))).to.be.rejectedWith(ExtractDocumentError, /empty content/);
+            });
+          });
+
+          context('with multiple selectors in one string', () => {
+            it('extracts content from the given HTML', async () => {
+              const result = await extract(new SourceDocument({
+                location: virtualLocation,
+                contentSelectors: 'h1, #link2',
+                content: rawHTML,
+              }));
+
+              expect(result).to.equal('Title\n=====\n\n[link 2](#anchor)');
+            });
           });
         });
 
-        context('with multiple selectors in one string', () => {
+        context('with an array of selectors', () => {
           it('extracts content from the given HTML', async () => {
-            const result = await extractFromHTML(new SourceDocument({
-              location: virtualLocation,
-              contentSelectors: 'h1, #link2',
+            const result = await extract(new SourceDocument({
               content: rawHTML,
+              location: virtualLocation,
+              contentSelectors: [ 'h1', '#link2' ],
             }));
 
             expect(result).to.equal('Title\n=====\n\n[link 2](#anchor)');
           });
-        });
-      });
 
-      context('with an array of selectors', () => {
-        it('extracts content from the given HTML', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: [ 'h1', '#link2' ],
-          }));
+          context('when one selector is dependent on another', () => {
+            it('extracts content from the given HTML', async () => {
+              const result = await extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: [ 'h1', 'h1 ~ p' ],
+              }));
 
-          expect(result).to.equal('Title\n=====\n\n[link 2](#anchor)');
+              expect(result).to.equal('Title\n=====\n\n[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)\n\n[link 3](http://absolute.url/link)\n\n[link 5](http://[INVALID_URL=http://www.example.org/)');
+            });
+          });
         });
 
-        context('when one selector is dependent on another', () => {
-          it('extracts content from the given HTML', async () => {
-            const result = await extractFromHTML(new SourceDocument({
-              content: rawHTML,
-              location: virtualLocation,
-              contentSelectors: [ 'h1', 'h1 ~ p' ],
-            }));
+        context('with range selector', () => {
+          context('with startBefore and endBefore', () => {
+            it('extracts content from the given HTML', async () => {
+              const result = await extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: {
+                  startBefore: '#link1',
+                  endBefore: '#link2',
+                },
+              }));
 
-            expect(result).to.equal('Title\n=====\n\n[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)\n\n[link 3](http://absolute.url/link)');
+              expect(result).to.equal('[link 1](https://exemple.com/relative/link)');
+            });
           });
-        });
-      });
+          context('with startBefore and endAfter', () => {
+            it('extracts content from the given HTML', async () => {
+              const result = await extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: {
+                  startBefore: '#link2',
+                  endAfter: '#link2',
+                },
+              }));
 
-      context('with range selector', () => {
-        context('with startBefore and endBefore', () => {
-          it('extracts content from the given HTML', async () => {
-            const result = await extractFromHTML(new SourceDocument({
-              content: rawHTML,
-              location: virtualLocation,
-              contentSelectors: {
-                startBefore: '#link1',
-                endBefore: '#link2',
-              },
-            }));
+              expect(result).to.equal('[link 2](#anchor)');
+            });
+          });
+          context('with startAfter and endBefore', () => {
+            it('extracts content from the given HTML', async () => {
+              const result = await extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: {
+                  startAfter: '#link1',
+                  endBefore: '#link3',
+                },
+              }));
 
-            expect(result).to.equal('[link 1](https://exemple.com/relative/link)');
+              expect(result).to.equal('[link 2](#anchor)');
+            });
           });
-        });
-        context('with startBefore and endAfter', () => {
-          it('extracts content from the given HTML', async () => {
-            const result = await extractFromHTML(new SourceDocument({
-              content: rawHTML,
-              location: virtualLocation,
-              contentSelectors: {
-                startBefore: '#link2',
-                endAfter: '#link2',
-              },
-            }));
+          context('with startAfter and endAfter', () => {
+            it('extracts content from the given HTML', async () => {
+              const result = await extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: {
+                  startAfter: '#link2',
+                  endAfter: '#link3',
+                },
+              }));
 
-            expect(result).to.equal('[link 2](#anchor)');
+              expect(result).to.equal('[link 3](http://absolute.url/link)');
+            });
+          });
+          context('with a "start" selector that has no match', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: {
+                  startAfter: '#paragraph1',
+                  endAfter: '#link2',
+                },
+              }))).to.be.rejectedWith(ExtractDocumentError, /"start" selector has no match/);
+            });
+          });
+          context('with an "end" selector that has no match', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: {
+                  startAfter: '#link2',
+                  endAfter: '#paragraph1',
+                },
+              }))).to.be.rejectedWith(ExtractDocumentError, /"end" selector has no match/);
+            });
           });
         });
-        context('with startAfter and endBefore', () => {
+
+        context('with an array of range selectors', () => {
           it('extracts content from the given HTML', async () => {
-            const result = await extractFromHTML(new SourceDocument({
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
-              contentSelectors: {
-                startAfter: '#link1',
-                endBefore: '#link3',
-              },
+              contentSelectors: [
+                {
+                  startAfter: '#link1',
+                  endAfter: '#link2',
+                },
+                {
+                  startAfter: '#link2',
+                  endAfter: '#link3',
+                },
+              ],
             }));
 
-            expect(result).to.equal('[link 2](#anchor)');
+            expect(result).to.equal('[link 2](#anchor)\n\n[link 3](http://absolute.url/link)');
           });
         });
-        context('with startAfter and endAfter', () => {
+
+        context('with an array of mixed string selectors and range selectors', () => {
           it('extracts content from the given HTML', async () => {
-            const result = await extractFromHTML(new SourceDocument({
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
-              contentSelectors: {
-                startAfter: '#link2',
-                endAfter: '#link3',
-              },
+              contentSelectors: [
+                'h1',
+                {
+                  startAfter: '#link2',
+                  endAfter: '#link3',
+                },
+              ],
             }));
 
-            expect(result).to.equal('[link 3](http://absolute.url/link)');
+            expect(result).to.equal('Title\n=====\n\n[link 3](http://absolute.url/link)');
           });
         });
-        context('with a "start" selector that has no match', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
+      });
+
+      describe('Remove', () => {
+        context('with a simple selector', () => {
+          it('removes the specified elements', async () => {
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
-              contentSelectors: {
-                startAfter: '#paragraph1',
-                endAfter: '#link2',
-              },
-            }))).to.be.rejectedWith(InaccessibleContentError, /"start" selector has no match/);
+              contentSelectors: 'body',
+              insignificantContentSelectors: 'h1',
+            }));
+
+            expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)\n\n[link 3](http://absolute.url/link)\n\n[link 5](http://[INVALID_URL=http://www.example.org/)');
           });
         });
-        context('with an "end" selector that has no match', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
+
+        context('with an array of string selectors', () => {
+          it('removes the specified elements', async () => {
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
-              contentSelectors: {
-                startAfter: '#link2',
-                endAfter: '#paragraph1',
-              },
-            }))).to.be.rejectedWith(InaccessibleContentError, /"end" selector has no match/);
-          });
-        });
-      });
-
-      context('with an array of range selectors', () => {
-        it('extracts content from the given HTML', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: [
-              {
-                startAfter: '#link1',
-                endAfter: '#link2',
-              },
-              {
-                startAfter: '#link2',
-                endAfter: '#link3',
-              },
-            ],
-          }));
-
-          expect(result).to.equal('[link 2](#anchor)\n\n[link 3](http://absolute.url/link)');
-        });
-      });
-
-      context('with an array of mixed string selectors and range selectors', () => {
-        it('extracts content from the given HTML', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: [
-              'h1',
-              {
-                startAfter: '#link2',
-                endAfter: '#link3',
-              },
-            ],
-          }));
-
-          expect(result).to.equal('Title\n=====\n\n[link 3](http://absolute.url/link)');
-        });
-      });
-    });
-
-    describe('Remove', () => {
-      context('with a simple selector', () => {
-        it('removes the specified elements', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            insignificantContentSelectors: 'h1',
-          }));
-
-          expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)\n\n[link 3](http://absolute.url/link)');
-        });
-      });
-
-      context('with an array of string selectors', () => {
-        it('removes the specified elements', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            insignificantContentSelectors: [ 'h1', '#link3' ],
-          }));
+              contentSelectors: 'body',
+              insignificantContentSelectors: [ 'h1', '#link3', '#link5' ],
+            }));
 
-          expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)');
+            expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)');
+          });
         });
-      });
 
-      context('with a simple range selector', () => {
-        it('removes the specified elements', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            insignificantContentSelectors: {
-              startBefore: '#link1',
-              endAfter: '#link3',
-            },
-          }));
-
-          expect(result).to.equal('Title\n=====');
-        });
-        context('with a "start" selector that has no match', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
+        context('with a simple range selector', () => {
+          it('removes the specified elements', async () => {
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
               contentSelectors: 'body',
               insignificantContentSelectors: {
-                startAfter: '#paragraph1',
-                endAfter: '#link2',
+                startBefore: '#link1',
+                endAfter: '#link5',
               },
-            }))).to.be.rejectedWith(InaccessibleContentError, /"start" selector has no match/);
+            }));
+
+            expect(result).to.equal('Title\n=====');
+          });
+          context('with a "start" selector that has no match', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: 'body',
+                insignificantContentSelectors: {
+                  startAfter: '#paragraph1',
+                  endAfter: '#link2',
+                },
+              }))).to.be.rejectedWith(ExtractDocumentError, /"start" selector has no match/);
+            });
+          });
+          context('with an "end" selector that has no match', () => {
+            it('throws an ExtractDocumentError error', async () => {
+              await expect(extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: 'body',
+                insignificantContentSelectors: {
+                  startAfter: '#link2',
+                  endAfter: '#paragraph1',
+                },
+              }))).to.be.rejectedWith(ExtractDocumentError, /"end" selector has no match/);
+            });
           });
         });
-        context('with an "end" selector that has no match', () => {
-          it('throws an InaccessibleContentError error', async () => {
-            await expect(extractFromHTML(new SourceDocument({
+        context('with an array of range selectors', () => {
+          it('removes all the selections', async () => {
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
               contentSelectors: 'body',
-              insignificantContentSelectors: {
-                startAfter: '#link2',
-                endAfter: '#paragraph1',
-              },
-            }))).to.be.rejectedWith(InaccessibleContentError, /"end" selector has no match/);
-          });
-        });
-      });
-      context('with an array of range selectors', () => {
-        it('removes all the selections', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            insignificantContentSelectors: [
-              {
-                startBefore: 'h1',
-                endBefore: '#link1',
-              },
-              {
-                startBefore: '#link3',
-                endAfter: '#link3',
-              },
-            ],
-          }));
-
-          expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)');
-        });
-      });
-
-      context('with an array of mixed selectors and range selectors', () => {
-        it('removes all the selections', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            insignificantContentSelectors: [
-              'h1',
-              {
-                startBefore: '#link3',
-                endAfter: '#link3',
-              },
-            ],
-          }));
+              insignificantContentSelectors: [
+                {
+                  startBefore: 'h1',
+                  endBefore: '#link1',
+                },
+                {
+                  startBefore: '#link3',
+                  endAfter: '#link5',
+                },
+              ],
+            }));
 
-          expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)');
+            expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)');
+          });
         });
 
-        context('where one selector is dependent on another', () => {
+        context('with an array of mixed selectors and range selectors', () => {
           it('removes all the selections', async () => {
-            const result = await extractFromHTML(new SourceDocument({
+            const result = await extract(new SourceDocument({
               content: rawHTML,
               location: virtualLocation,
               contentSelectors: 'body',
               insignificantContentSelectors: [
                 'h1',
                 {
-                  startAfter: 'h1',
-                  endBefore: '#link2',
+                  startBefore: '#link3',
+                  endAfter: '#link5',
                 },
               ],
             }));
 
-            expect(result).to.equal('[link 2](#anchor)\n\n[link 3](http://absolute.url/link)');
+            expect(result).to.equal('[link 1](https://exemple.com/relative/link)\n\n[link 2](#anchor)');
+          });
+
+          context('where one selector is dependent on another', () => {
+            it('removes all the selections', async () => {
+              const result = await extract(new SourceDocument({
+                content: rawHTML,
+                location: virtualLocation,
+                contentSelectors: 'body',
+                insignificantContentSelectors: [
+                  'h1',
+                  {
+                    startAfter: 'h1',
+                    endBefore: '#link2',
+                  },
+                ],
+              }));
+
+              expect(result).to.equal('[link 2](#anchor)\n\n[link 3](http://absolute.url/link)\n\n[link 5](http://[INVALID_URL=http://www.example.org/)');
+            });
           });
         });
       });
-    });
 
-    describe('Filter', () => {
-      context('with a synchronous filter', () => {
-        it('extracts content from the given HTML also with given additional filter', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            filters: [additionalFilter.removeLinks],
-          }));
-
-          expect(result).to.equal(expectedExtractedWithAdditional);
+      describe('Filter', () => {
+        context('with a synchronous filter', () => {
+          it('extracts content from the given HTML also with given additional filter', async () => {
+            const result = await extract(new SourceDocument({
+              content: rawHTML,
+              location: virtualLocation,
+              contentSelectors: 'body',
+              filters: [additionalFilter.removeLinks],
+            }));
+
+            expect(result).to.equal(expectedExtractedWithAdditional);
+          });
         });
-      });
 
-      context('with an asynchronous filter', () => {
-        it('extracts content from the given HTML also with given additional filter', async () => {
-          const result = await extractFromHTML(new SourceDocument({
-            content: rawHTML,
-            location: virtualLocation,
-            contentSelectors: 'body',
-            filters: [additionalFilter.removeLinksAsync],
-          }));
+        context('with an asynchronous filter', () => {
+          it('extracts content from the given HTML also with given additional filter', async () => {
+            const result = await extract(new SourceDocument({
+              content: rawHTML,
+              location: virtualLocation,
+              contentSelectors: 'body',
+              filters: [additionalFilter.removeLinksAsync],
+            }));
 
-          expect(result).to.equal(expectedExtractedWithAdditional);
+            expect(result).to.equal(expectedExtractedWithAdditional);
+          });
         });
       });
     });
-  });
 
-  describe('#extractFromPDF', () => {
-    let pdfContent;
-    let expectedExtractedContent;
+    context('from PDF content', () => {
+      let pdfContent;
+      let expectedExtractedContent;
 
-    before(async () => {
-      pdfContent = await fs.readFile(path.resolve(__dirname, '../../../test/fixtures/terms.pdf'));
-      expectedExtractedContent = await fs.readFile(
-        path.resolve(__dirname, '../../../test/fixtures/termsFromPDF.md'),
-        { encoding: 'utf8' },
-      );
-    });
+      before(async () => {
+        pdfContent = await fs.readFile(path.resolve(__dirname, '../../../test/fixtures/terms.pdf'));
+        expectedExtractedContent = await fs.readFile(
+          path.resolve(__dirname, '../../../test/fixtures/termsFromPDF.md'),
+          { encoding: 'utf8' },
+        );
+      });
 
-    it('extracts content from the given PDF', async () => {
-      expect(await extractFromPDF({ content: pdfContent })).to.equal(expectedExtractedContent);
+      it('extracts content from the given PDF', async () => {
+        expect(await extract({ content: pdfContent, mimeType: mime.getType('pdf') })).to.equal(expectedExtractedContent);
+      });
+
+      context('when PDF contains no text', () => {
+        it('throws an ExtractDocumentError error', async () => {
+          await expect(extract({ content: await fs.readFile(path.resolve(__dirname, '../../../test/fixtures/termsNoText.pdf')), mimeType: mime.getType('pdf') })).to.be.rejectedWith(ExtractDocumentError, /contains no text/);
+        });
+      });
     });
   });
 });